On the Likely Utility of Hybrid Weights Optimized for Variances in Hybrid Error Covariance Models
NASA Astrophysics Data System (ADS)
Satterfield, E.; Hodyss, D.; Kuhl, D.; Bishop, C. H.
2017-12-01
Because of imperfections in ensemble data assimilation schemes, one cannot assume that the ensemble covariance is equal to the true error covariance of a forecast. Previous work demonstrated how information about the distribution of true error variances given an ensemble sample variance can be revealed from an archive of (observation-minus-forecast, ensemble-variance) data pairs. Here, we derive a simple and intuitively compelling formula to obtain the mean of this distribution of true error variances given an ensemble sample variance from (observation-minus-forecast, ensemble-variance) data pairs produced by a single run of a data assimilation system. This formula takes the form of a Hybrid weighted average of the climatological forecast error variance and the ensemble sample variance. Here, we test the extent to which these readily obtainable weights can be used to rapidly optimize the covariance weights used in Hybrid data assimilation systems that employ weighted averages of static covariance models and flow-dependent ensemble based covariance models. Univariate data assimilation and multi-variate cycling ensemble data assimilation are considered. In both cases, it is found that our computationally efficient formula gives Hybrid weights that closely approximate the optimal weights found through the simple but computationally expensive process of testing every plausible combination of weights.
Errors in radial velocity variance from Doppler wind lidar
DOE Office of Scientific and Technical Information (OSTI.GOV)
Wang, H.; Barthelmie, R. J.; Doubrawa, P.
A high-fidelity lidar turbulence measurement technique relies on accurate estimates of radial velocity variance that are subject to both systematic and random errors determined by the autocorrelation function of radial velocity, the sampling rate, and the sampling duration. Our paper quantifies the effect of the volumetric averaging in lidar radial velocity measurements on the autocorrelation function and the dependence of the systematic and random errors on the sampling duration, using both statistically simulated and observed data. For current-generation scanning lidars and sampling durations of about 30 min and longer, during which the stationarity assumption is valid for atmospheric flows, themore » systematic error is negligible but the random error exceeds about 10%.« less
Errors in radial velocity variance from Doppler wind lidar
Wang, H.; Barthelmie, R. J.; Doubrawa, P.; ...
2016-08-29
A high-fidelity lidar turbulence measurement technique relies on accurate estimates of radial velocity variance that are subject to both systematic and random errors determined by the autocorrelation function of radial velocity, the sampling rate, and the sampling duration. Our paper quantifies the effect of the volumetric averaging in lidar radial velocity measurements on the autocorrelation function and the dependence of the systematic and random errors on the sampling duration, using both statistically simulated and observed data. For current-generation scanning lidars and sampling durations of about 30 min and longer, during which the stationarity assumption is valid for atmospheric flows, themore » systematic error is negligible but the random error exceeds about 10%.« less
Weighting by Inverse Variance or by Sample Size in Random-Effects Meta-Analysis
ERIC Educational Resources Information Center
Marin-Martinez, Fulgencio; Sanchez-Meca, Julio
2010-01-01
Most of the statistical procedures in meta-analysis are based on the estimation of average effect sizes from a set of primary studies. The optimal weight for averaging a set of independent effect sizes is the inverse variance of each effect size, but in practice these weights have to be estimated, being affected by sampling error. When assuming a…
Evaluation and optimization of sampling errors for the Monte Carlo Independent Column Approximation
NASA Astrophysics Data System (ADS)
Räisänen, Petri; Barker, W. Howard
2004-07-01
The Monte Carlo Independent Column Approximation (McICA) method for computing domain-average broadband radiative fluxes is unbiased with respect to the full ICA, but its flux estimates contain conditional random noise. McICA's sampling errors are evaluated here using a global climate model (GCM) dataset and a correlated-k distribution (CKD) radiation scheme. Two approaches to reduce McICA's sampling variance are discussed. The first is to simply restrict all of McICA's samples to cloudy regions. This avoids wasting precious few samples on essentially homogeneous clear skies. Clear-sky fluxes need to be computed separately for this approach, but this is usually done in GCMs for diagnostic purposes anyway. Second, accuracy can be improved by repeated sampling, and averaging those CKD terms with large cloud radiative effects. Although this naturally increases computational costs over the standard CKD model, random errors for fluxes and heating rates are reduced by typically 50% to 60%, for the present radiation code, when the total number of samples is increased by 50%. When both variance reduction techniques are applied simultaneously, globally averaged flux and heating rate random errors are reduced by a factor of #3.
Estimation of sampling error uncertainties in observed surface air temperature change in China
NASA Astrophysics Data System (ADS)
Hua, Wei; Shen, Samuel S. P.; Weithmann, Alexander; Wang, Huijun
2017-08-01
This study examines the sampling error uncertainties in the monthly surface air temperature (SAT) change in China over recent decades, focusing on the uncertainties of gridded data, national averages, and linear trends. Results indicate that large sampling error variances appear at the station-sparse area of northern and western China with the maximum value exceeding 2.0 K2 while small sampling error variances are found at the station-dense area of southern and eastern China with most grid values being less than 0.05 K2. In general, the negative temperature existed in each month prior to the 1980s, and a warming in temperature began thereafter, which accelerated in the early and mid-1990s. The increasing trend in the SAT series was observed for each month of the year with the largest temperature increase and highest uncertainty of 0.51 ± 0.29 K (10 year)-1 occurring in February and the weakest trend and smallest uncertainty of 0.13 ± 0.07 K (10 year)-1 in August. The sampling error uncertainties in the national average annual mean SAT series are not sufficiently large to alter the conclusion of the persistent warming in China. In addition, the sampling error uncertainties in the SAT series show a clear variation compared with other uncertainty estimation methods, which is a plausible reason for the inconsistent variations between our estimate and other studies during this period.
Sampling design optimisation for rainfall prediction using a non-stationary geostatistical model
NASA Astrophysics Data System (ADS)
Wadoux, Alexandre M. J.-C.; Brus, Dick J.; Rico-Ramirez, Miguel A.; Heuvelink, Gerard B. M.
2017-09-01
The accuracy of spatial predictions of rainfall by merging rain-gauge and radar data is partly determined by the sampling design of the rain-gauge network. Optimising the locations of the rain-gauges may increase the accuracy of the predictions. Existing spatial sampling design optimisation methods are based on minimisation of the spatially averaged prediction error variance under the assumption of intrinsic stationarity. Over the past years, substantial progress has been made to deal with non-stationary spatial processes in kriging. Various well-documented geostatistical models relax the assumption of stationarity in the mean, while recent studies show the importance of considering non-stationarity in the variance for environmental processes occurring in complex landscapes. We optimised the sampling locations of rain-gauges using an extension of the Kriging with External Drift (KED) model for prediction of rainfall fields. The model incorporates both non-stationarity in the mean and in the variance, which are modelled as functions of external covariates such as radar imagery, distance to radar station and radar beam blockage. Spatial predictions are made repeatedly over time, each time recalibrating the model. The space-time averaged KED variance was minimised by Spatial Simulated Annealing (SSA). The methodology was tested using a case study predicting daily rainfall in the north of England for a one-year period. Results show that (i) the proposed non-stationary variance model outperforms the stationary variance model, and (ii) a small but significant decrease of the rainfall prediction error variance is obtained with the optimised rain-gauge network. In particular, it pays off to place rain-gauges at locations where the radar imagery is inaccurate, while keeping the distribution over the study area sufficiently uniform.
Prediction-error variance in Bayesian model updating: a comparative study
NASA Astrophysics Data System (ADS)
Asadollahi, Parisa; Li, Jian; Huang, Yong
2017-04-01
In Bayesian model updating, the likelihood function is commonly formulated by stochastic embedding in which the maximum information entropy probability model of prediction error variances plays an important role and it is Gaussian distribution subject to the first two moments as constraints. The selection of prediction error variances can be formulated as a model class selection problem, which automatically involves a trade-off between the average data-fit of the model class and the information it extracts from the data. Therefore, it is critical for the robustness in the updating of the structural model especially in the presence of modeling errors. To date, three ways of considering prediction error variances have been seem in the literature: 1) setting constant values empirically, 2) estimating them based on the goodness-of-fit of the measured data, and 3) updating them as uncertain parameters by applying Bayes' Theorem at the model class level. In this paper, the effect of different strategies to deal with the prediction error variances on the model updating performance is investigated explicitly. A six-story shear building model with six uncertain stiffness parameters is employed as an illustrative example. Transitional Markov Chain Monte Carlo is used to draw samples of the posterior probability density function of the structure model parameters as well as the uncertain prediction variances. The different levels of modeling uncertainty and complexity are modeled through three FE models, including a true model, a model with more complexity, and a model with modeling error. Bayesian updating is performed for the three FE models considering the three aforementioned treatments of the prediction error variances. The effect of number of measurements on the model updating performance is also examined in the study. The results are compared based on model class assessment and indicate that updating the prediction error variances as uncertain parameters at the model class level produces more robust results especially when the number of measurement is small.
Combining forecast weights: Why and how?
NASA Astrophysics Data System (ADS)
Yin, Yip Chee; Kok-Haur, Ng; Hock-Eam, Lim
2012-09-01
This paper proposes a procedure called forecast weight averaging which is a specific combination of forecast weights obtained from different methods of constructing forecast weights for the purpose of improving the accuracy of pseudo out of sample forecasting. It is found that under certain specified conditions, forecast weight averaging can lower the mean squared forecast error obtained from model averaging. In addition, we show that in a linear and homoskedastic environment, this superior predictive ability of forecast weight averaging holds true irrespective whether the coefficients are tested by t statistic or z statistic provided the significant level is within the 10% range. By theoretical proofs and simulation study, we have shown that model averaging like, variance model averaging, simple model averaging and standard error model averaging, each produces mean squared forecast error larger than that of forecast weight averaging. Finally, this result also holds true marginally when applied to business and economic empirical data sets, Gross Domestic Product (GDP growth rate), Consumer Price Index (CPI) and Average Lending Rate (ALR) of Malaysia.
Santin-Janin, Hugues; Hugueny, Bernard; Aubry, Philippe; Fouchet, David; Gimenez, Olivier; Pontier, Dominique
2014-01-01
Data collected to inform time variations in natural population size are tainted by sampling error. Ignoring sampling error in population dynamics models induces bias in parameter estimators, e.g., density-dependence. In particular, when sampling errors are independent among populations, the classical estimator of the synchrony strength (zero-lag correlation) is biased downward. However, this bias is rarely taken into account in synchrony studies although it may lead to overemphasizing the role of intrinsic factors (e.g., dispersal) with respect to extrinsic factors (the Moran effect) in generating population synchrony as well as to underestimating the extinction risk of a metapopulation. The aim of this paper was first to illustrate the extent of the bias that can be encountered in empirical studies when sampling error is neglected. Second, we presented a space-state modelling approach that explicitly accounts for sampling error when quantifying population synchrony. Third, we exemplify our approach with datasets for which sampling variance (i) has been previously estimated, and (ii) has to be jointly estimated with population synchrony. Finally, we compared our results to those of a standard approach neglecting sampling variance. We showed that ignoring sampling variance can mask a synchrony pattern whatever its true value and that the common practice of averaging few replicates of population size estimates poorly performed at decreasing the bias of the classical estimator of the synchrony strength. The state-space model used in this study provides a flexible way of accurately quantifying the strength of synchrony patterns from most population size data encountered in field studies, including over-dispersed count data. We provided a user-friendly R-program and a tutorial example to encourage further studies aiming at quantifying the strength of population synchrony to account for uncertainty in population size estimates.
Santin-Janin, Hugues; Hugueny, Bernard; Aubry, Philippe; Fouchet, David; Gimenez, Olivier; Pontier, Dominique
2014-01-01
Background Data collected to inform time variations in natural population size are tainted by sampling error. Ignoring sampling error in population dynamics models induces bias in parameter estimators, e.g., density-dependence. In particular, when sampling errors are independent among populations, the classical estimator of the synchrony strength (zero-lag correlation) is biased downward. However, this bias is rarely taken into account in synchrony studies although it may lead to overemphasizing the role of intrinsic factors (e.g., dispersal) with respect to extrinsic factors (the Moran effect) in generating population synchrony as well as to underestimating the extinction risk of a metapopulation. Methodology/Principal findings The aim of this paper was first to illustrate the extent of the bias that can be encountered in empirical studies when sampling error is neglected. Second, we presented a space-state modelling approach that explicitly accounts for sampling error when quantifying population synchrony. Third, we exemplify our approach with datasets for which sampling variance (i) has been previously estimated, and (ii) has to be jointly estimated with population synchrony. Finally, we compared our results to those of a standard approach neglecting sampling variance. We showed that ignoring sampling variance can mask a synchrony pattern whatever its true value and that the common practice of averaging few replicates of population size estimates poorly performed at decreasing the bias of the classical estimator of the synchrony strength. Conclusion/Significance The state-space model used in this study provides a flexible way of accurately quantifying the strength of synchrony patterns from most population size data encountered in field studies, including over-dispersed count data. We provided a user-friendly R-program and a tutorial example to encourage further studies aiming at quantifying the strength of population synchrony to account for uncertainty in population size estimates. PMID:24489839
Wonnapinij, Passorn; Chinnery, Patrick F.; Samuels, David C.
2010-01-01
In cases of inherited pathogenic mitochondrial DNA (mtDNA) mutations, a mother and her offspring generally have large and seemingly random differences in the amount of mutated mtDNA that they carry. Comparisons of measured mtDNA mutation level variance values have become an important issue in determining the mechanisms that cause these large random shifts in mutation level. These variance measurements have been made with samples of quite modest size, which should be a source of concern because higher-order statistics, such as variance, are poorly estimated from small sample sizes. We have developed an analysis of the standard error of variance from a sample of size n, and we have defined error bars for variance measurements based on this standard error. We calculate variance error bars for several published sets of measurements of mtDNA mutation level variance and show how the addition of the error bars alters the interpretation of these experimental results. We compare variance measurements from human clinical data and from mouse models and show that the mutation level variance is clearly higher in the human data than it is in the mouse models at both the primary oocyte and offspring stages of inheritance. We discuss how the standard error of variance can be used in the design of experiments measuring mtDNA mutation level variance. Our results show that variance measurements based on fewer than 20 measurements are generally unreliable and ideally more than 50 measurements are required to reliably compare variances with less than a 2-fold difference. PMID:20362273
NASA Technical Reports Server (NTRS)
Chelton, Dudley B.; Schlax, Michael G.
1991-01-01
The sampling error of an arbitrary linear estimate of a time-averaged quantity constructed from a time series of irregularly spaced observations at a fixed located is quantified through a formalism. The method is applied to satellite observations of chlorophyll from the coastal zone color scanner. The two specific linear estimates under consideration are the composite average formed from the simple average of all observations within the averaging period and the optimal estimate formed by minimizing the mean squared error of the temporal average based on all the observations in the time series. The resulting suboptimal estimates are shown to be more accurate than composite averages. Suboptimal estimates are also found to be nearly as accurate as optimal estimates using the correct signal and measurement error variances and correlation functions for realistic ranges of these parameters, which makes it a viable practical alternative to the composite average method generally employed at present.
Smooth empirical Bayes estimation of observation error variances in linear systems
NASA Technical Reports Server (NTRS)
Martz, H. F., Jr.; Lian, M. W.
1972-01-01
A smooth empirical Bayes estimator was developed for estimating the unknown random scale component of each of a set of observation error variances. It is shown that the estimator possesses a smaller average squared error loss than other estimators for a discrete time linear system.
Assumption-free estimation of the genetic contribution to refractive error across childhood.
Guggenheim, Jeremy A; St Pourcain, Beate; McMahon, George; Timpson, Nicholas J; Evans, David M; Williams, Cathy
2015-01-01
Studies in relatives have generally yielded high heritability estimates for refractive error: twins 75-90%, families 15-70%. However, because related individuals often share a common environment, these estimates are inflated (via misallocation of unique/common environment variance). We calculated a lower-bound heritability estimate for refractive error free from such bias. Between the ages 7 and 15 years, participants in the Avon Longitudinal Study of Parents and Children (ALSPAC) underwent non-cycloplegic autorefraction at regular research clinics. At each age, an estimate of the variance in refractive error explained by single nucleotide polymorphism (SNP) genetic variants was calculated using genome-wide complex trait analysis (GCTA) using high-density genome-wide SNP genotype information (minimum N at each age=3,404). The variance in refractive error explained by the SNPs ("SNP heritability") was stable over childhood: Across age 7-15 years, SNP heritability averaged 0.28 (SE=0.08, p<0.001). The genetic correlation for refractive error between visits varied from 0.77 to 1.00 (all p<0.001) demonstrating that a common set of SNPs was responsible for the genetic contribution to refractive error across this period of childhood. Simulations suggested lack of cycloplegia during autorefraction led to a small underestimation of SNP heritability (adjusted SNP heritability=0.35; SE=0.09). To put these results in context, the variance in refractive error explained (or predicted) by the time participants spent outdoors was <0.005 and by the time spent reading was <0.01, based on a parental questionnaire completed when the child was aged 8-9 years old. Genetic variation captured by common SNPs explained approximately 35% of the variation in refractive error between unrelated subjects. This value sets an upper limit for predicting refractive error using existing SNP genotyping arrays, although higher-density genotyping in larger samples and inclusion of interaction effects is expected to raise this figure toward twin- and family-based heritability estimates. The same SNPs influenced refractive error across much of childhood. Notwithstanding the strong evidence of association between time outdoors and myopia, and time reading and myopia, less than 1% of the variance in myopia at age 15 was explained by crude measures of these two risk factors, indicating that their effects may be limited, at least when averaged over the whole population.
Generalized Variance Function Applications in Forestry
James Alegria; Charles T. Scott; Charles T. Scott
1991-01-01
Adequately predicting the sampling errors of tabular data can reduce printing costs by eliminating the need to publish separate sampling error tables. Two generalized variance functions (GVFs) found in the literature and three GVFs derived for this study were evaluated for their ability to predict the sampling error of tabular forestry estimates. The recommended GVFs...
Analysis of Darwin Rainfall Data: Implications on Sampling Strategy
NASA Technical Reports Server (NTRS)
Rafael, Qihang Li; Bras, Rafael L.; Veneziano, Daniele
1996-01-01
Rainfall data collected by radar in the vicinity of Darwin, Australia, have been analyzed in terms of their mean, variance, autocorrelation of area-averaged rain rate, and diurnal variation. It is found that, when compared with the well-studied GATE (Global Atmospheric Research Program Atlantic Tropical Experiment) data, Darwin rainfall has larger coefficient of variation (CV), faster reduction of CV with increasing area size, weaker temporal correlation, and a strong diurnal cycle and intermittence. The coefficient of variation for Darwin rainfall has larger magnitude and exhibits larger spatial variability over the sea portion than over the land portion within the area of radar coverage. Stationary, and nonstationary models have been used to study the sampling errors associated with space-based rainfall measurement. The nonstationary model shows that the sampling error is sensitive to the starting sampling time for some sampling frequencies, due to the diurnal cycle of rain, but not for others. Sampling experiments using data also show such sensitivity. When the errors are averaged over starting time, the results of the experiments and the stationary and nonstationary models match each other very closely. In the small areas for which data are available for I>oth Darwin and GATE, the sampling error is expected to be larger for Darwin due to its larger CV.
ERIC Educational Resources Information Center
Richerson, Lindsay P.; Watkins, Marley W.; Beaujean, A. Alexander
2014-01-01
Measurement invariance of the Wechsler Intelligence Scale for Children--Fourth Edition (WISC-IV) was investigated with a group of 352 students eligible for psychoeducational evaluations tested, on average, 2.8 years apart. Configural, metric, and scalar invariance were found. However, the error variance of the Coding subtest was not constant…
Xu, Chonggang; Gertner, George
2013-01-01
Fourier Amplitude Sensitivity Test (FAST) is one of the most popular uncertainty and sensitivity analysis techniques. It uses a periodic sampling approach and a Fourier transformation to decompose the variance of a model output into partial variances contributed by different model parameters. Until now, the FAST analysis is mainly confined to the estimation of partial variances contributed by the main effects of model parameters, but does not allow for those contributed by specific interactions among parameters. In this paper, we theoretically show that FAST analysis can be used to estimate partial variances contributed by both main effects and interaction effects of model parameters using different sampling approaches (i.e., traditional search-curve based sampling, simple random sampling and random balance design sampling). We also analytically calculate the potential errors and biases in the estimation of partial variances. Hypothesis tests are constructed to reduce the effect of sampling errors on the estimation of partial variances. Our results show that compared to simple random sampling and random balance design sampling, sensitivity indices (ratios of partial variances to variance of a specific model output) estimated by search-curve based sampling generally have higher precision but larger underestimations. Compared to simple random sampling, random balance design sampling generally provides higher estimation precision for partial variances contributed by the main effects of parameters. The theoretical derivation of partial variances contributed by higher-order interactions and the calculation of their corresponding estimation errors in different sampling schemes can help us better understand the FAST method and provide a fundamental basis for FAST applications and further improvements. PMID:24143037
Xu, Chonggang; Gertner, George
2011-01-01
Fourier Amplitude Sensitivity Test (FAST) is one of the most popular uncertainty and sensitivity analysis techniques. It uses a periodic sampling approach and a Fourier transformation to decompose the variance of a model output into partial variances contributed by different model parameters. Until now, the FAST analysis is mainly confined to the estimation of partial variances contributed by the main effects of model parameters, but does not allow for those contributed by specific interactions among parameters. In this paper, we theoretically show that FAST analysis can be used to estimate partial variances contributed by both main effects and interaction effects of model parameters using different sampling approaches (i.e., traditional search-curve based sampling, simple random sampling and random balance design sampling). We also analytically calculate the potential errors and biases in the estimation of partial variances. Hypothesis tests are constructed to reduce the effect of sampling errors on the estimation of partial variances. Our results show that compared to simple random sampling and random balance design sampling, sensitivity indices (ratios of partial variances to variance of a specific model output) estimated by search-curve based sampling generally have higher precision but larger underestimations. Compared to simple random sampling, random balance design sampling generally provides higher estimation precision for partial variances contributed by the main effects of parameters. The theoretical derivation of partial variances contributed by higher-order interactions and the calculation of their corresponding estimation errors in different sampling schemes can help us better understand the FAST method and provide a fundamental basis for FAST applications and further improvements.
Non-stationary internal tides observed with satellite altimetry
NASA Astrophysics Data System (ADS)
Ray, R. D.; Zaron, E. D.
2011-09-01
Temporal variability of the internal tide is inferred from a 17-year combined record of Topex/Poseidon and Jason satellite altimeters. A global sampling of along-track sea-surface height wavenumber spectra finds that non-stationary variance is generally 25% or less of the average variance at wavenumbers characteristic of mode-1 tidal internal waves. With some exceptions the non-stationary variance does not exceed 0.25 cm2. The mode-2 signal, where detectable, contains a larger fraction of non-stationary variance, typically 50% or more. Temporal subsetting of the data reveals interannual variability barely significant compared with tidal estimation error from 3-year records. Comparison of summer vs. winter conditions shows only one region of noteworthy seasonal changes, the northern South China Sea. Implications for the anticipated SWOT altimeter mission are briefly discussed.
Non-Stationary Internal Tides Observed with Satellite Altimetry
NASA Technical Reports Server (NTRS)
Ray, Richard D.; Zaron, E. D.
2011-01-01
Temporal variability of the internal tide is inferred from a 17-year combined record of Topex/Poseidon and Jason satellite altimeters. A global sampling of along-track sea-surface height wavenumber spectra finds that non-stationary variance is generally 25% or less of the average variance at wavenumbers characteristic of mode-l tidal internal waves. With some exceptions the non-stationary variance does not exceed 0.25 sq cm. The mode-2 signal, where detectable, contains a larger fraction of non-stationary variance, typically 50% or more. Temporal subsetting of the data reveals interannual variability barely significant compared with tidal estimation error from 3-year records. Comparison of summer vs. winter conditions shows only one region of noteworthy seasonal changes, the northern South China Sea. Implications for the anticipated SWOT altimeter mission are briefly discussed.
Holmes, John B; Dodds, Ken G; Lee, Michael A
2017-03-02
An important issue in genetic evaluation is the comparability of random effects (breeding values), particularly between pairs of animals in different contemporary groups. This is usually referred to as genetic connectedness. While various measures of connectedness have been proposed in the literature, there is general agreement that the most appropriate measure is some function of the prediction error variance-covariance matrix. However, obtaining the prediction error variance-covariance matrix is computationally demanding for large-scale genetic evaluations. Many alternative statistics have been proposed that avoid the computational cost of obtaining the prediction error variance-covariance matrix, such as counts of genetic links between contemporary groups, gene flow matrices, and functions of the variance-covariance matrix of estimated contemporary group fixed effects. In this paper, we show that a correction to the variance-covariance matrix of estimated contemporary group fixed effects will produce the exact prediction error variance-covariance matrix averaged by contemporary group for univariate models in the presence of single or multiple fixed effects and one random effect. We demonstrate the correction for a series of models and show that approximations to the prediction error matrix based solely on the variance-covariance matrix of estimated contemporary group fixed effects are inappropriate in certain circumstances. Our method allows for the calculation of a connectedness measure based on the prediction error variance-covariance matrix by calculating only the variance-covariance matrix of estimated fixed effects. Since the number of fixed effects in genetic evaluation is usually orders of magnitudes smaller than the number of random effect levels, the computational requirements for our method should be reduced.
Toward Joint Hypothesis-Tests Seismic Event Screening Analysis: Ms|mb and Event Depth
DOE Office of Scientific and Technical Information (OSTI.GOV)
Anderson, Dale; Selby, Neil
2012-08-14
Well established theory can be used to combine single-phenomenology hypothesis tests into a multi-phenomenology event screening hypothesis test (Fisher's and Tippett's tests). Commonly used standard error in Ms:mb event screening hypothesis test is not fully consistent with physical basis. Improved standard error - Better agreement with physical basis, and correctly partitions error to include Model Error as a component of variance, correctly reduces station noise variance through network averaging. For 2009 DPRK test - Commonly used standard error 'rejects' H0 even with better scaling slope ({beta} = 1, Selby et al.), improved standard error 'fails to rejects' H0.
A new statistic to express the uncertainty of kriging predictions for purposes of survey planning.
NASA Astrophysics Data System (ADS)
Lark, R. M.; Lapworth, D. J.
2014-05-01
It is well-known that one advantage of kriging for spatial prediction is that, given the random effects model, the prediction error variance can be computed a priori for alternative sampling designs. This allows one to compare sampling schemes, in particular sampling at different densities, and so to decide on one which meets requirements in terms of the uncertainty of the resulting predictions. However, the planning of sampling schemes must account not only for statistical considerations, but also logistics and cost. This requires effective communication between statisticians, soil scientists and data users/sponsors such as managers, regulators or civil servants. In our experience the latter parties are not necessarily able to interpret the prediction error variance as a measure of uncertainty for decision making. In some contexts (particularly the solution of very specific problems at large cartographic scales, e.g. site remediation and precision farming) it is possible to translate uncertainty of predictions into a loss function directly comparable with the cost incurred in increasing precision. Often, however, sampling must be planned for more generic purposes (e.g. baseline or exploratory geochemical surveys). In this latter context the prediction error variance may be of limited value to a non-statistician who has to make a decision on sample intensity and associated cost. We propose an alternative criterion for these circumstances to aid communication between statisticians and data users about the uncertainty of geostatistical surveys based on different sampling intensities. The criterion is the consistency of estimates made from two non-coincident instantiations of a proposed sample design. We consider square sample grids, one instantiation is offset from the second by half the grid spacing along the rows and along the columns. If a sample grid is coarse relative to the important scales of variation in the target property then the consistency of predictions from two instantiations is expected to be small, and can be increased by reducing the grid spacing. The measure of consistency is the correlation between estimates from the two instantiations of the sample grid, averaged over a grid cell. We call this the offset correlation, it can be calculated from the variogram. We propose that this measure is easier to grasp intuitively than the prediction error variance, and has the advantage of having an upper bound (1.0) which will aid its interpretation. This quality measure is illustrated for some hypothetical examples, considering both ordinary kriging and factorial kriging of the variable of interest. It is also illustrated using data on metal concentrations in the soil of north-east England.
Quantizing and sampling considerations in digital phased-locked loops
NASA Technical Reports Server (NTRS)
Hurst, G. T.; Gupta, S. C.
1974-01-01
The quantizer problem is first considered. The conditions under which the uniform white sequence model for the quantizer error is valid are established independent of the sampling rate. An equivalent spectral density is defined for the quantizer error resulting in an effective SNR value. This effective SNR may be used to determine quantized performance from infinitely fine quantized results. Attention is given to sampling rate considerations. Sampling rate characteristics of the digital phase-locked loop (DPLL) structure are investigated for the infinitely fine quantized system. The predicted phase error variance equation is examined as a function of the sampling rate. Simulation results are presented and a method is described which enables the minimum required sampling rate to be determined from the predicted phase error variance equations.
Eaton, Jeffrey W.; Bao, Le
2017-01-01
Objectives The aim of the study was to propose and demonstrate an approach to allow additional nonsampling uncertainty about HIV prevalence measured at antenatal clinic sentinel surveillance (ANC-SS) in model-based inferences about trends in HIV incidence and prevalence. Design Mathematical model fitted to surveillance data with Bayesian inference. Methods We introduce a variance inflation parameter σinfl2 that accounts for the uncertainty of nonsampling errors in ANC-SS prevalence. It is additive to the sampling error variance. Three approaches are tested for estimating σinfl2 using ANC-SS and household survey data from 40 subnational regions in nine countries in sub-Saharan, as defined in UNAIDS 2016 estimates. Methods were compared using in-sample fit and out-of-sample prediction of ANC-SS data, fit to household survey prevalence data, and the computational implications. Results Introducing the additional variance parameter σinfl2 increased the error variance around ANC-SS prevalence observations by a median of 2.7 times (interquartile range 1.9–3.8). Using only sampling error in ANC-SS prevalence ( σinfl2=0), coverage of 95% prediction intervals was 69% in out-of-sample prediction tests. This increased to 90% after introducing the additional variance parameter σinfl2. The revised probabilistic model improved model fit to household survey prevalence and increased epidemic uncertainty intervals most during the early epidemic period before 2005. Estimating σinfl2 did not increase the computational cost of model fitting. Conclusions: We recommend estimating nonsampling error in ANC-SS as an additional parameter in Bayesian inference using the Estimation and Projection Package model. This approach may prove useful for incorporating other data sources such as routine prevalence from Prevention of mother-to-child transmission testing into future epidemic estimates. PMID:28296801
Kirkpatrick, Robert M; McGue, Matt; Iacono, William G
2015-03-01
The present study of general cognitive ability attempts to replicate and extend previous investigations of a biometric moderator, family-of-origin socioeconomic status (SES), in a sample of 2,494 pairs of adolescent twins, non-twin biological siblings, and adoptive siblings assessed with individually administered IQ tests. We hypothesized that SES would covary positively with additive-genetic variance and negatively with shared-environmental variance. Important potential confounds unaddressed in some past studies, such as twin-specific effects, assortative mating, and differential heritability by trait level, were found to be negligible. In our main analysis, we compared models by their sample-size corrected AIC, and base our statistical inference on model-averaged point estimates and standard errors. Additive-genetic variance increased with SES-an effect that was statistically significant and robust to model specification. We found no evidence that SES moderated shared-environmental influence. We attempt to explain the inconsistent replication record of these effects, and provide suggestions for future research.
Kirkpatrick, Robert M.; McGue, Matt; Iacono, William G.
2015-01-01
The present study of general cognitive ability attempts to replicate and extend previous investigations of a biometric moderator, family-of-origin socioeconomic status (SES), in a sample of 2,494 pairs of adolescent twins, non-twin biological siblings, and adoptive siblings assessed with individually administered IQ tests. We hypothesized that SES would covary positively with additive-genetic variance and negatively with shared-environmental variance. Important potential confounds unaddressed in some past studies, such as twin-specific effects, assortative mating, and differential heritability by trait level, were found to be negligible. In our main analysis, we compared models by their sample-size corrected AIC, and base our statistical inference on model-averaged point estimates and standard errors. Additive-genetic variance increased with SES—an effect that was statistically significant and robust to model specification. We found no evidence that SES moderated shared-environmental influence. We attempt to explain the inconsistent replication record of these effects, and provide suggestions for future research. PMID:25539975
NASA Technical Reports Server (NTRS)
Deloach, Richard; Obara, Clifford J.; Goodman, Wesley L.
2012-01-01
This paper documents a check standard wind tunnel test conducted in the Langley 0.3-Meter Transonic Cryogenic Tunnel (0.3M TCT) that was designed and analyzed using the Modern Design of Experiments (MDOE). The test designed to partition the unexplained variance of typical wind tunnel data samples into two constituent components, one attributable to ordinary random error, and one attributable to systematic error induced by covariate effects. Covariate effects in wind tunnel testing are discussed, with examples. The impact of systematic (non-random) unexplained variance on the statistical independence of sequential measurements is reviewed. The corresponding correlation among experimental errors is discussed, as is the impact of such correlation on experimental results generally. The specific experiment documented herein was organized as a formal test for the presence of unexplained variance in representative samples of wind tunnel data, in order to quantify the frequency with which such systematic error was detected, and its magnitude relative to ordinary random error. Levels of systematic and random error reported here are representative of those quantified in other facilities, as cited in the references.
An Empirical State Error Covariance Matrix for Batch State Estimation
NASA Technical Reports Server (NTRS)
Frisbee, Joseph H., Jr.
2011-01-01
State estimation techniques serve effectively to provide mean state estimates. However, the state error covariance matrices provided as part of these techniques suffer from some degree of lack of confidence in their ability to adequately describe the uncertainty in the estimated states. A specific problem with the traditional form of state error covariance matrices is that they represent only a mapping of the assumed observation error characteristics into the state space. Any errors that arise from other sources (environment modeling, precision, etc.) are not directly represented in a traditional, theoretical state error covariance matrix. Consider that an actual observation contains only measurement error and that an estimated observation contains all other errors, known and unknown. It then follows that a measurement residual (the difference between expected and observed measurements) contains all errors for that measurement. Therefore, a direct and appropriate inclusion of the actual measurement residuals in the state error covariance matrix will result in an empirical state error covariance matrix. This empirical state error covariance matrix will fully account for the error in the state estimate. By way of a literal reinterpretation of the equations involved in the weighted least squares estimation algorithm, it is possible to arrive at an appropriate, and formally correct, empirical state error covariance matrix. The first specific step of the method is to use the average form of the weighted measurement residual variance performance index rather than its usual total weighted residual form. Next it is helpful to interpret the solution to the normal equations as the average of a collection of sample vectors drawn from a hypothetical parent population. From here, using a standard statistical analysis approach, it directly follows as to how to determine the standard empirical state error covariance matrix. This matrix will contain the total uncertainty in the state estimate, regardless as to the source of the uncertainty. Also, in its most straight forward form, the technique only requires supplemental calculations to be added to existing batch algorithms. The generation of this direct, empirical form of the state error covariance matrix is independent of the dimensionality of the observations. Mixed degrees of freedom for an observation set are allowed. As is the case with any simple, empirical sample variance problems, the presented approach offers an opportunity (at least in the case of weighted least squares) to investigate confidence interval estimates for the error covariance matrix elements. The diagonal or variance terms of the error covariance matrix have a particularly simple form to associate with either a multiple degree of freedom chi-square distribution (more approximate) or with a gamma distribution (less approximate). The off diagonal or covariance terms of the matrix are less clear in their statistical behavior. However, the off diagonal covariance matrix elements still lend themselves to standard confidence interval error analysis. The distributional forms associated with the off diagonal terms are more varied and, perhaps, more approximate than those associated with the diagonal terms. Using a simple weighted least squares sample problem, results obtained through use of the proposed technique are presented. The example consists of a simple, two observer, triangulation problem with range only measurements. Variations of this problem reflect an ideal case (perfect knowledge of the range errors) and a mismodeled case (incorrect knowledge of the range errors).
Bezdjian, Serena; Tuvblad, Catherine; Wang, Pan; Raine, Adrian; Baker, Laura A
2014-11-01
In the present study, we investigated genetic and environmental effects on motor impulsivity from childhood to late adolescence using a longitudinal sample of twins from ages 9 to 18 years. Motor impulsivity was assessed using errors of commission (no-go errors) in a visual go/no-go task at 4 time points: ages 9-10, 11-13, 14-15, and 16-18 years. Significant genetic and nonshared environmental effects on motor impulsivity were found at each of the 4 waves of assessment with genetic factors explaining 22%-41% of the variance within each of the 4 waves. Phenotypically, children's average performance improved across age (i.e., fewer no-go errors during later assessments). Multivariate biometric analyses revealed that common genetic factors influenced 12%-40% of the variance in motor impulsivity across development, whereas nonshared environmental factors common to all time points contributed to 2%-52% of the variance. Nonshared environmental influences specific to each time point also significantly influenced motor impulsivity. Overall, results demonstrated that although genetic factors were critical to motor impulsivity across development, both common and specific nonshared environmental factors played a strong role in the development of motor impulsivity across age. (PsycINFO Database Record (c) 2014 APA, all rights reserved).
Statistics of the radiated field of a space-to-earth microwave power transfer system
NASA Technical Reports Server (NTRS)
Stevens, G. H.; Leininger, G.
1976-01-01
Statistics such as average power density pattern, variance of the power density pattern and variance of the beam pointing error are related to hardware parameters such as transmitter rms phase error and rms amplitude error. Also a limitation on spectral width of the phase reference for phase control was established. A 1 km diameter transmitter appears feasible provided the total rms insertion phase errors of the phase control modules does not exceed 10 deg, amplitude errors do not exceed 10% rms, and the phase reference spectral width does not exceed approximately 3 kHz. With these conditions the expected radiation pattern is virtually the same as the error free pattern, and the rms beam pointing error would be insignificant (approximately 10 meters).
USDA-ARS?s Scientific Manuscript database
We proposed a method to estimate the error variance among non-replicated genotypes, thus to estimate the genetic parameters by using replicated controls. We derived formulas to estimate sampling variances of the genetic parameters. Computer simulation indicated that the proposed methods of estimatin...
Comparative test on several forms of background error covariance in 3DVar
NASA Astrophysics Data System (ADS)
Shao, Aimei
2013-04-01
The background error covariance matrix (Hereinafter referred to as B matrix) plays an important role in the three-dimensional variational (3DVar) data assimilation method. However, it is difficult to get B matrix accurately because true atmospheric state is unknown. Therefore, some methods were developed to estimate B matrix (e.g. NMC method, innovation analysis method, recursive filters, and ensemble method such as EnKF). Prior to further development and application of these methods, the function of several B matrixes estimated by these methods in 3Dvar is worth studying and evaluating. For this reason, NCEP reanalysis data and forecast data are used to test the effectiveness of the several B matrixes with VAF (Huang, 1999) method. Here the NCEP analysis is treated as the truth and in this case the forecast error is known. The data from 2006 to 2007 is used as the samples to estimate B matrix and the data in 2008 is used to verify the assimilation effects. The 48h and 24h forecast valid at the same time is used to estimate B matrix with NMC method. B matrix can be represented by a correlation part (a non-diagonal matrix) and a variance part (a diagonal matrix of variances). Gaussian filter function as an approximate approach is used to represent the variation of correlation coefficients with distance in numerous 3DVar systems. On the basis of the assumption, the following several forms of B matrixes are designed and test with VAF in the comparative experiments: (1) error variance and the characteristic lengths are fixed and setted to their mean value averaged over the analysis domain; (2) similar to (1), but the mean characteristic lengths reduce to 50 percent for the height and 60 percent for the temperature of the original; (3) similar to (2), but error variance calculated directly by the historical data is space-dependent; (4) error variance and characteristic lengths are all calculated directly by the historical data; (5) B matrix is estimated directly by the historical data; (6) similar to (5), but a localization process is performed; (7) B matrix is estimated by NMC method but error variance is reduced by 1.7 times in order that the value is close to that calculated from the true forecast error samples; (8) similar to (7), but the localization similar to (6) is performed. Experimental results with the different B matrixes show that for the Gaussian-type B matrix the characteristic lengths calculated from the true error samples don't bring a good analysis results. However, the reduced characteristic lengths (about half of the original one) can lead to a good analysis. If the B matrix estimated directly from the historical data is used in 3DVar, the assimilation effect can not reach to the best. The better assimilation results are generated with the application of reduced characteristic length and localization. Even so, it hasn't obvious advantage compared with Gaussian-type B matrix with the optimal characteristic length. It implies that the Gaussian-type B matrix, widely used for operational 3DVar system, can get a good analysis with the appropriate characteristic lengths. The crucial problem is how to determine the appropriate characteristic lengths. (This work is supported by the National Natural Science Foundation of China (41275102, 40875063), and the Fundamental Research Funds for the Central Universities (lzujbky-2010-9) )
Evaluation of TRMM Ground-Validation Radar-Rain Errors Using Rain Gauge Measurements
NASA Technical Reports Server (NTRS)
Wang, Jianxin; Wolff, David B.
2009-01-01
Ground-validation (GV) radar-rain products are often utilized for validation of the Tropical Rainfall Measuring Mission (TRMM) spaced-based rain estimates, and hence, quantitative evaluation of the GV radar-rain product error characteristics is vital. This study uses quality-controlled gauge data to compare with TRMM GV radar rain rates in an effort to provide such error characteristics. The results show that significant differences of concurrent radar-gauge rain rates exist at various time scales ranging from 5 min to 1 day, despite lower overall long-term bias. However, the differences between the radar area-averaged rain rates and gauge point rain rates cannot be explained as due to radar error only. The error variance separation method is adapted to partition the variance of radar-gauge differences into the gauge area-point error variance and radar rain estimation error variance. The results provide relatively reliable quantitative uncertainty evaluation of TRMM GV radar rain estimates at various times scales, and are helpful to better understand the differences between measured radar and gauge rain rates. It is envisaged that this study will contribute to better utilization of GV radar rain products to validate versatile spaced-based rain estimates from TRMM, as well as the proposed Global Precipitation Measurement, and other satellites.
Efficiently estimating salmon escapement uncertainty using systematically sampled data
Reynolds, Joel H.; Woody, Carol Ann; Gove, Nancy E.; Fair, Lowell F.
2007-01-01
Fish escapement is generally monitored using nonreplicated systematic sampling designs (e.g., via visual counts from towers or hydroacoustic counts). These sampling designs support a variety of methods for estimating the variance of the total escapement. Unfortunately, all the methods give biased results, with the magnitude of the bias being determined by the underlying process patterns. Fish escapement commonly exhibits positive autocorrelation and nonlinear patterns, such as diurnal and seasonal patterns. For these patterns, poor choice of variance estimator can needlessly increase the uncertainty managers have to deal with in sustaining fish populations. We illustrate the effect of sampling design and variance estimator choice on variance estimates of total escapement for anadromous salmonids from systematic samples of fish passage. Using simulated tower counts of sockeye salmon Oncorhynchus nerka escapement on the Kvichak River, Alaska, five variance estimators for nonreplicated systematic samples were compared to determine the least biased. Using the least biased variance estimator, four confidence interval estimators were compared for expected coverage and mean interval width. Finally, five systematic sampling designs were compared to determine the design giving the smallest average variance estimate for total annual escapement. For nonreplicated systematic samples of fish escapement, all variance estimators were positively biased. Compared to the other estimators, the least biased estimator reduced bias by, on average, from 12% to 98%. All confidence intervals gave effectively identical results. Replicated systematic sampling designs consistently provided the smallest average estimated variance among those compared.
ERIC Educational Resources Information Center
Vardeman, Stephen B.; Wendelberger, Joanne R.
2005-01-01
There is a little-known but very simple generalization of the standard result that for uncorrelated random variables with common mean [mu] and variance [sigma][superscript 2], the expected value of the sample variance is [sigma][superscript 2]. The generalization justifies the use of the usual standard error of the sample mean in possibly…
ERIC Educational Resources Information Center
Oranje, Andreas
2006-01-01
A multitude of methods has been proposed to estimate the sampling variance of ratio estimates in complex samples (Wolter, 1985). Hansen and Tepping (1985) studied some of those variance estimators and found that a high coefficient of variation (CV) of the denominator of a ratio estimate is indicative of a biased estimate of the standard error of a…
Triple collocation based merging of satellite soil moisture retrievals
USDA-ARS?s Scientific Manuscript database
We propose a method for merging soil moisture retrievals from space borne active and passive microwave instruments based on weighted averaging taking into account the error characteristics of the individual data sets. The merging scheme is parameterized using error variance estimates obtained from u...
Experimental study on an FBG strain sensor
NASA Astrophysics Data System (ADS)
Liu, Hong-lin; Zhu, Zheng-wei; Zheng, Yong; Liu, Bang; Xiao, Feng
2018-01-01
Landslides and other geological disasters occur frequently and often cause high financial and humanitarian cost. The real-time, early-warning monitoring of landslides has important significance in reducing casualties and property losses. In this paper, by taking the high initial precision and high sensitivity advantage of FBG, an FBG strain sensor is designed combining FBGs with inclinometer. The sensor was regarded as a cantilever beam with one end fixed. According to the anisotropic material properties of the inclinometer, a theoretical formula between the FBG wavelength and the deflection of the sensor was established using the elastic mechanics principle. Accuracy of the formula established had been verified through laboratory calibration testing and model slope monitoring experiments. The displacement of landslide could be calculated by the established theoretical formula using the changing values of FBG central wavelength obtained by the demodulation instrument remotely. Results showed that the maximum error at different heights was 9.09%; the average of the maximum error was 6.35%, and its corresponding variance was 2.12; the minimum error was 4.18%; the average of the minimum error was 5.99%, and its corresponding variance was 0.50. The maximum error of the theoretical and the measured displacement decrease gradually, and the variance of the error also decreases gradually. This indicates that the theoretical results are more and more reliable. It also shows that the sensor and the theoretical formula established in this paper can be used for remote, real-time, high precision and early warning monitoring of the slope.
Incorporating measurement error in n = 1 psychological autoregressive modeling.
Schuurman, Noémi K; Houtveen, Jan H; Hamaker, Ellen L
2015-01-01
Measurement error is omnipresent in psychological data. However, the vast majority of applications of autoregressive time series analyses in psychology do not take measurement error into account. Disregarding measurement error when it is present in the data results in a bias of the autoregressive parameters. We discuss two models that take measurement error into account: An autoregressive model with a white noise term (AR+WN), and an autoregressive moving average (ARMA) model. In a simulation study we compare the parameter recovery performance of these models, and compare this performance for both a Bayesian and frequentist approach. We find that overall, the AR+WN model performs better. Furthermore, we find that for realistic (i.e., small) sample sizes, psychological research would benefit from a Bayesian approach in fitting these models. Finally, we illustrate the effect of disregarding measurement error in an AR(1) model by means of an empirical application on mood data in women. We find that, depending on the person, approximately 30-50% of the total variance was due to measurement error, and that disregarding this measurement error results in a substantial underestimation of the autoregressive parameters.
Support, shape and number of replicate samples for tree foliage analysis.
Luyssaert, Sebastiaan; Mertens, Jan; Raitio, Hannu
2003-06-01
Many fundamental features of a sampling program are determined by the heterogeneity of the object under study and the settings for the error (alpha), the power (beta), the effect size (ES), the number of replicate samples, and sample support, which is a feature that is often overlooked. The number of replicates, alpha, beta, ES, and sample support are interconnected. The effect of the sample support and its shape on the required number of replicate samples was investigated by means of a resampling method. The method was applied to a simulated distribution of Cd in the crown of a Salix fragilis L. tree. Increasing the dimensions of the sample support results in a decrease in the variance of the element concentration under study. Analysis of the variance is often the foundation of statistical tests, therefore, valid statistical testing requires the use of a fixed sample support during the experiment. This requirement might be difficult to meet in time-series analyses and long-term monitoring programs. Sample supports have their largest dimension in the direction with the largest heterogeneity, i.e. the direction representing the crown height, and this will give more accurate results than supports with other shapes. Taking the relationships between the sample support and the variance of the element concentrations in tree crowns into account provides guidelines for sampling efficiency in terms of precision and costs. In terms of time, the optimal support to test whether the average Cd concentration of the crown exceeds a threshold value is 0.405 m3 (alpha = 0.05, beta = 0.20, ES = 1.0 mg kg(-1) dry mass). The average weight of this support is 23 g dry mass, and 11 replicate samples need to be taken. It should be noted that in this case the optimal support applies to Cd under conditions similar to those of the simulation, but not necessarily all the examinations for this tree species, element, and hypothesis test.
Austin, Peter C
2016-12-30
Propensity score methods are used to reduce the effects of observed confounding when using observational data to estimate the effects of treatments or exposures. A popular method of using the propensity score is inverse probability of treatment weighting (IPTW). When using this method, a weight is calculated for each subject that is equal to the inverse of the probability of receiving the treatment that was actually received. These weights are then incorporated into the analyses to minimize the effects of observed confounding. Previous research has found that these methods result in unbiased estimation when estimating the effect of treatment on survival outcomes. However, conventional methods of variance estimation were shown to result in biased estimates of standard error. In this study, we conducted an extensive set of Monte Carlo simulations to examine different methods of variance estimation when using a weighted Cox proportional hazards model to estimate the effect of treatment. We considered three variance estimation methods: (i) a naïve model-based variance estimator; (ii) a robust sandwich-type variance estimator; and (iii) a bootstrap variance estimator. We considered estimation of both the average treatment effect and the average treatment effect in the treated. We found that the use of a bootstrap estimator resulted in approximately correct estimates of standard errors and confidence intervals with the correct coverage rates. The other estimators resulted in biased estimates of standard errors and confidence intervals with incorrect coverage rates. Our simulations were informed by a case study examining the effect of statin prescribing on mortality. © 2016 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd. © 2016 The Authors. Statistics in Medicine published by John Wiley & Sons Ltd.
Standard errors in forest area
Joseph McCollum
2002-01-01
I trace the development of standard error equations for forest area, beginning with the theory behind double sampling and the variance of a product. The discussion shifts to the particular problem of forest area - at which time the theory becomes relevant. There are subtle difficulties in figuring out which variance of a product equation should be used. The equations...
An internal pilot design for prospective cancer screening trials with unknown disease prevalence.
Brinton, John T; Ringham, Brandy M; Glueck, Deborah H
2015-10-13
For studies that compare the diagnostic accuracy of two screening tests, the sample size depends on the prevalence of disease in the study population, and on the variance of the outcome. Both parameters may be unknown during the design stage, which makes finding an accurate sample size difficult. To solve this problem, we propose adapting an internal pilot design. In this adapted design, researchers will accrue some percentage of the planned sample size, then estimate both the disease prevalence and the variances of the screening tests. The updated estimates of the disease prevalence and variance are used to conduct a more accurate power and sample size calculation. We demonstrate that in large samples, the adapted internal pilot design produces no Type I inflation. For small samples (N less than 50), we introduce a novel adjustment of the critical value to control the Type I error rate. We apply the method to two proposed prospective cancer screening studies: 1) a small oral cancer screening study in individuals with Fanconi anemia and 2) a large oral cancer screening trial. Conducting an internal pilot study without adjusting the critical value can cause Type I error rate inflation in small samples, but not in large samples. An internal pilot approach usually achieves goal power and, for most studies with sample size greater than 50, requires no Type I error correction. Further, we have provided a flexible and accurate approach to bound Type I error below a goal level for studies with small sample size.
Variance of discharge estimates sampled using acoustic Doppler current profilers from moving boats
Garcia, Carlos M.; Tarrab, Leticia; Oberg, Kevin; Szupiany, Ricardo; Cantero, Mariano I.
2012-01-01
This paper presents a model for quantifying the random errors (i.e., variance) of acoustic Doppler current profiler (ADCP) discharge measurements from moving boats for different sampling times. The model focuses on the random processes in the sampled flow field and has been developed using statistical methods currently available for uncertainty analysis of velocity time series. Analysis of field data collected using ADCP from moving boats from three natural rivers of varying sizes and flow conditions shows that, even though the estimate of the integral time scale of the actual turbulent flow field is larger than the sampling interval, the integral time scale of the sampled flow field is on the order of the sampling interval. Thus, an equation for computing the variance error in discharge measurements associated with different sampling times, assuming uncorrelated flow fields is appropriate. The approach is used to help define optimal sampling strategies by choosing the exposure time required for ADCPs to accurately measure flow discharge.
Within-Tunnel Variations in Pressure Data for Three Transonic Wind Tunnels
NASA Technical Reports Server (NTRS)
DeLoach, Richard
2014-01-01
This paper compares the results of pressure measurements made on the same test article with the same test matrix in three transonic wind tunnels. A comparison is presented of the unexplained variance associated with polar replicates acquired in each tunnel. The impact of a significance component of systematic (not random) unexplained variance is reviewed, and the results of analyses of variance are presented to assess the degree of significant systematic error in these representative wind tunnel tests. Total uncertainty estimates are reported for 140 samples of pressure data, quantifying the effects of within-polar random errors and between-polar systematic bias errors.
Effects of low sampling rate in the digital data-transition tracking loop
NASA Technical Reports Server (NTRS)
Mileant, A.; Million, S.; Hinedi, S.
1994-01-01
This article describes the performance of the all-digital data-transition tracking loop (DTTL) with coherent and noncoherent sampling using nonlinear theory. The effects of few samples per symbol and of noncommensurate sampling and symbol rates are addressed and analyzed. Their impact on the probability density and variance of the phase error are quantified through computer simulations. It is shown that the performance of the all-digital DTTL approaches its analog counterpart when the sampling and symbol rates are noncommensurate (i.e., the number of samples per symbol is an irrational number). The loop signal-to-noise ratio (SNR) (inverse of phase error variance) degrades when the number of samples per symbol is an odd integer but degrades even further for even integers.
How many drinks did you have on September 11, 2001?
Perrine, M W Bud; Schroder, Kerstin E E
2005-07-01
This study tested the predictability of error in retrospective self-reports of alcohol consumption on September 11, 2001, among 80 Vermont light, medium and heavy drinkers. Subjects were 52 men and 28 women participating in daily self-reports of alcohol consumption for a total of 2 years, collected via interactive voice response technology (IVR). In addition, retrospective self-reports of alcohol consumption on September 11, 2001, were collected by telephone interview 4-5 days following the terrorist attacks. Retrospective error was calculated as the difference between the IVR self-report of drinking behavior on September 11 and the retrospective self-report collected by telephone interview. Retrospective error was analyzed as a function of gender and baseline drinking behavior during the 365 days preceding September 11, 2001 (termed "the baseline"). The intraclass correlation (ICC) between daily IVR and retrospective self-reports of alcohol consumption on September 11 was .80. Women provided, on average, more accurate self-reports (ICC = .96) than men (ICC = .72) but displayed more underreporting bias in retrospective responses. Amount and individual variability of alcohol consumption during the 1-year baseline explained, on average, 11% of the variance in overreporting (r = .33), 9% of the variance in underreporting (r = .30) and 25% of the variance in the overall magnitude of error (r = .50), with correlations up to .62 (r2 = .38). The size and direction of error were clearly predictable from the amount and variation in drinking behavior during the 1-year baseline period. The results demonstrate the utility and detail of information that can be derived from daily IVR self-reports in the analysis of retrospective error.
Fischer, A; Friggens, N C; Berry, D P; Faverdin, P
2018-07-01
The ability to properly assess and accurately phenotype true differences in feed efficiency among dairy cows is key to the development of breeding programs for improving feed efficiency. The variability among individuals in feed efficiency is commonly characterised by the residual intake approach. Residual feed intake is represented by the residuals of a linear regression of intake on the corresponding quantities of the biological functions that consume (or release) energy. However, the residuals include both, model fitting and measurement errors as well as any variability in cow efficiency. The objective of this study was to isolate the individual animal variability in feed efficiency from the residual component. Two separate models were fitted, in one the standard residual energy intake (REI) was calculated as the residual of a multiple linear regression of lactation average net energy intake (NEI) on lactation average milk energy output, average metabolic BW, as well as lactation loss and gain of body condition score. In the other, a linear mixed model was used to simultaneously fit fixed linear regressions and random cow levels on the biological traits and intercept using fortnight repeated measures for the variables. This method split the predicted NEI in two parts: one quantifying the population mean intercept and coefficients, and one quantifying cow-specific deviations in the intercept and coefficients. The cow-specific part of predicted NEI was assumed to isolate true differences in feed efficiency among cows. NEI and associated energy expenditure phenotypes were available for the first 17 fortnights of lactation from 119 Holstein cows; all fed a constant energy-rich diet. Mixed models fitting cow-specific intercept and coefficients to different combinations of the aforementioned energy expenditure traits, calculated on a fortnightly basis, were compared. The variance of REI estimated with the lactation average model represented only 8% of the variance of measured NEI. Among all compared mixed models, the variance of the cow-specific part of predicted NEI represented between 53% and 59% of the variance of REI estimated from the lactation average model or between 4% and 5% of the variance of measured NEI. The remaining 41% to 47% of the variance of REI estimated with the lactation average model may therefore reflect model fitting errors or measurement errors. In conclusion, the use of a mixed model framework with cow-specific random regressions seems to be a promising method to isolate the cow-specific component of REI in dairy cows.
Trattner, Sigal; Prinsen, Peter; Wiegert, Jens; Gerland, Elazar-Lars; Shefer, Efrat; Morton, Tom; Thompson, Carla M; Yagil, Yoad; Cheng, Bin; Jambawalikar, Sachin; Al-Senan, Rani; Amurao, Maxwell; Halliburton, Sandra S; Einstein, Andrew J
2017-12-01
Metal-oxide-semiconductor field-effect transistors (MOSFETs) serve as a helpful tool for organ radiation dosimetry and their use has grown in computed tomography (CT). While different approaches have been used for MOSFET calibration, those using the commonly available 100 mm pencil ionization chamber have not incorporated measurements performed throughout its length, and moreover, no previous work has rigorously evaluated the multiple sources of error involved in MOSFET calibration. In this paper, we propose a new MOSFET calibration approach to translate MOSFET voltage measurements into absorbed dose from CT, based on serial measurements performed throughout the length of a 100-mm ionization chamber, and perform an analysis of the errors of MOSFET voltage measurements and four sources of error in calibration. MOSFET calibration was performed at two sites, to determine single calibration factors for tube potentials of 80, 100, and 120 kVp, using a 100-mm-long pencil ion chamber and a cylindrical computed tomography dose index (CTDI) phantom of 32 cm diameter. The dose profile along the 100-mm ion chamber axis was sampled in 5 mm intervals by nine MOSFETs in the nine holes of the CTDI phantom. Variance of the absorbed dose was modeled as a sum of the MOSFET voltage measurement variance and the calibration factor variance, the latter being comprised of three main subcomponents: ionization chamber reading variance, MOSFET-to-MOSFET variation and a contribution related to the fact that the average calibration factor of a few MOSFETs was used as an estimate for the average value of all MOSFETs. MOSFET voltage measurement error was estimated based on sets of repeated measurements. The calibration factor overall voltage measurement error was calculated from the above analysis. Calibration factors determined were close to those reported in the literature and by the manufacturer (~3 mV/mGy), ranging from 2.87 to 3.13 mV/mGy. The error σ V of a MOSFET voltage measurement was shown to be proportional to the square root of the voltage V: σV=cV where c = 0.11 mV. A main contributor to the error in the calibration factor was the ionization chamber reading error with 5% error. The usage of a single calibration factor for all MOSFETs introduced an additional error of about 5-7%, depending on the number of MOSFETs that were used to determine the single calibration factor. The expected overall error in a high-dose region (~30 mGy) was estimated to be about 8%, compared to 6% when an individual MOSFET calibration was performed. For a low-dose region (~3 mGy), these values were 13% and 12%. A MOSFET calibration method was developed using a 100-mm pencil ion chamber and a CTDI phantom, accompanied by an absorbed dose error analysis reflecting multiple sources of measurement error. When using a single calibration factor, per tube potential, for different MOSFETs, only a small error was introduced into absorbed dose determinations, thus supporting the use of a single calibration factor for experiments involving many MOSFETs, such as those required to accurately estimate radiation effective dose. © 2017 American Association of Physicists in Medicine.
Oita, Azusa; Tsuboi, Yuuri; Date, Yasuhiro; Oshima, Takahiro; Sakata, Kenji; Yokoyama, Akiko; Moriya, Shigeharu; Kikuchi, Jun
2018-04-24
There is an increasing need for assessing aquatic ecosystems that are globally endangered. Since aquatic ecosystems are complex, integrated consideration of multiple factors utilizing omics technologies can help us better understand aquatic ecosystems. An integrated strategy linking three analytical (machine learning, factor mapping, and forecast-error-variance decomposition) approaches for extracting the features of surface water from datasets comprising ions, metabolites, and microorganisms is proposed herein. The three developed approaches can be employed for diverse datasets of sample sizes and experimentally analyzed factors. The three approaches are applied to explore the features of bay water surrounding Odaiba, Tokyo, Japan, as a case study. Firstly, the machine learning approach separated 681 surface water samples within Japan into three clusters, categorizing Odaiba water into seawater with relatively low inorganic ions, including Mg, Ba, and B. Secondly, the factor mapping approach illustrated Odaiba water samples from the summer as rich in multiple amino acids and some other metabolites and poor in inorganic ions relative to other seasons based on their seasonal dynamics. Finally, forecast-error-variance decomposition using vector autoregressive models indicated that a type of microalgae (Raphidophyceae) grows in close correlation with alanine, succinic acid, and valine on filters and with isobutyric acid and 4-hydroxybenzoic acid in filtrate, Ba, and average wind speed. Our integrated strategy can be used to examine many biological, chemical, and environmental physical factors to analyze surface water. Copyright © 2018. Published by Elsevier B.V.
Estimation of lipids and lean mass of migrating sandpipers
Skagen, Susan K.; Knopf, Fritz L.; Cade, Brian S.
1993-01-01
Estimation of lean mass and lipid levels in birds involves the derivation of predictive equations that relate morphological measurements and, more recently, total body electrical conductivity (TOBEC) indices to known lean and lipid masses. Using cross-validation techniques, we evaluated the ability of several published and new predictive equations to estimate lean and lipid mass of Semipalmated Sandpipers (Calidris pusilla) and White-rumped Sandpipers (C. fuscicollis). We also tested ideas of Morton et al. (1991), who stated that current statistical approaches to TOBEC methodology misrepresent precision in estimating body fat. Three published interspecific equations using TOBEC indices predicted lean and lipid masses of our sample of birds with average errors of 8-28% and 53-155%, respectively. A new two-species equation relating lean mass and TOBEC indices revealed average errors of 4.6% and 23.2% in predicting lean and lipid mass, respectively. New intraspecific equations that estimate lipid mass directly from body mass, morphological measurements, and TOBEC indices yielded about a 13% error in lipid estimates. Body mass and morphological measurements explained a substantial portion of the variance (about 90%) in fat mass of both species. Addition of TOBEC indices improved the predictive model more for the smaller than for the larger sandpiper. TOBEC indices explained an additional 7.8% and 2.6% of the variance in fat mass and reduced the minimum breadth of prediction intervals by 0.95 g (32%) and 0.39 g (13%) for Semipalmated and White-rumped Sandpipers, respectively. The breadth of prediction intervals for models used to predict fat levels of individual birds must be considered when interpreting the resultant lipid estimates.
Incorporating measurement error in n = 1 psychological autoregressive modeling
Schuurman, Noémi K.; Houtveen, Jan H.; Hamaker, Ellen L.
2015-01-01
Measurement error is omnipresent in psychological data. However, the vast majority of applications of autoregressive time series analyses in psychology do not take measurement error into account. Disregarding measurement error when it is present in the data results in a bias of the autoregressive parameters. We discuss two models that take measurement error into account: An autoregressive model with a white noise term (AR+WN), and an autoregressive moving average (ARMA) model. In a simulation study we compare the parameter recovery performance of these models, and compare this performance for both a Bayesian and frequentist approach. We find that overall, the AR+WN model performs better. Furthermore, we find that for realistic (i.e., small) sample sizes, psychological research would benefit from a Bayesian approach in fitting these models. Finally, we illustrate the effect of disregarding measurement error in an AR(1) model by means of an empirical application on mood data in women. We find that, depending on the person, approximately 30–50% of the total variance was due to measurement error, and that disregarding this measurement error results in a substantial underestimation of the autoregressive parameters. PMID:26283988
Tangen, C M; Koch, G G
1999-03-01
In the randomized clinical trial setting, controlling for covariates is expected to produce variance reduction for the treatment parameter estimate and to adjust for random imbalances of covariates between the treatment groups. However, for the logistic regression model, variance reduction is not obviously obtained. This can lead to concerns about the assumptions of the logistic model. We introduce a complementary nonparametric method for covariate adjustment. It provides results that are usually compatible with expectations for analysis of covariance. The only assumptions required are based on randomization and sampling arguments. The resulting treatment parameter is a (unconditional) population average log-odds ratio that has been adjusted for random imbalance of covariates. Data from a randomized clinical trial are used to compare results from the traditional maximum likelihood logistic method with those from the nonparametric logistic method. We examine treatment parameter estimates, corresponding standard errors, and significance levels in models with and without covariate adjustment. In addition, we discuss differences between unconditional population average treatment parameters and conditional subpopulation average treatment parameters. Additional features of the nonparametric method, including stratified (multicenter) and multivariate (multivisit) analyses, are illustrated. Extensions of this methodology to the proportional odds model are also made.
Test analysis and research on static choice reaction ability of commercial vehicle drivers
NASA Astrophysics Data System (ADS)
Zhang, Lingchao; Wei, Lang; Qiao, Jie; Tian, Shun; Wang, Shengchang
2017-03-01
Drivers' choice reaction ability has a certain relation with safe driving. It has important significance to research its influence on traffic safety. Firstly, the paper uses a choice reaction detector developed by research group to detect drivers' choice reaction ability of commercial vehicles, and gets 2641 effective samples. Then by using mathematical statistics method, the paper founds that average reaction time from accident group has no difference with non-accident group, and then introduces a variance rate of reaction time as a new index to replace it. The result shows that the test index choice reaction errors and variance rate of reaction time have positive correlations with accidents. Finally, according to testing results of the detector, the paper formulates a detection threshold with four levels for helping transportation companies to assess commercial vehicles drivers.
Kogut, Katherine; Eisen, Ellen A.; Jewell, Nicholas P.; Quirós-Alcalá, Lesliam; Castorina, Rosemary; Chevrier, Jonathan; Holland, Nina T.; Barr, Dana Boyd; Kavanagh-Baird, Geri; Eskenazi, Brenda
2012-01-01
Background: Dialkyl phosphate (DAP) metabolites in spot urine samples are frequently used to characterize children’s exposures to organophosphorous (OP) pesticides. However, variable exposure and short biological half-lives of OP pesticides could result in highly variable measurements, leading to exposure misclassification. Objective: We examined within- and between-child variability in DAP metabolites in urine samples collected during 1 week. Methods: We collected spot urine samples over 7 consecutive days from 25 children (3–6 years of age). On two of the days, we collected 24-hr voids. We assessed the reproducibility of urinary DAP metabolite concentrations and evaluated the sensitivity and specificity of spot urine samples as predictors of high (top 20%) or elevated (top 40%) weekly average DAP metabolite concentrations. Results: Within-child variance exceeded between-child variance by a factor of two to eight, depending on metabolite grouping. Although total DAP concentrations in single spot urine samples were moderately to strongly associated with concentrations in same-day 24-hr samples (r ≈ 0.6–0.8, p < 0.01), concentrations in spot samples collected > 1 day apart and in 24-hr samples collected 3 days apart were weakly correlated (r ≈ –0.21 to 0.38). Single spot samples predicted high (top 20%) and elevated (top 40%) full-week average total DAP excretion with only moderate sensitivity (≈ 0.52 and ≈ 0.67, respectively) but relatively high specificity (≈ 0.88 and ≈ 0.78, respectively). Conclusions: The high variability we observed in children’s DAP metabolite concentrations suggests that single-day urine samples provide only a brief snapshot of exposure. Sensitivity analyses suggest that classification of cumulative OP exposure based on spot samples is prone to type 2 classification errors. PMID:23052012
On the impact of relatedness on SNP association analysis.
Gross, Arnd; Tönjes, Anke; Scholz, Markus
2017-12-06
When testing for SNP (single nucleotide polymorphism) associations in related individuals, observations are not independent. Simple linear regression assuming independent normally distributed residuals results in an increased type I error and the power of the test is also affected in a more complicate manner. Inflation of type I error is often successfully corrected by genomic control. However, this reduces the power of the test when relatedness is of concern. In the present paper, we derive explicit formulae to investigate how heritability and strength of relatedness contribute to variance inflation of the effect estimate of the linear model. Further, we study the consequences of variance inflation on hypothesis testing and compare the results with those of genomic control correction. We apply the developed theory to the publicly available HapMap trio data (N=129), the Sorbs (a self-contained population with N=977 characterised by a cryptic relatedness structure) and synthetic family studies with different sample sizes (ranging from N=129 to N=999) and different degrees of relatedness. We derive explicit and easily to apply approximation formulae to estimate the impact of relatedness on the variance of the effect estimate of the linear regression model. Variance inflation increases with increasing heritability. Relatedness structure also impacts the degree of variance inflation as shown for example family structures. Variance inflation is smallest for HapMap trios, followed by a synthetic family study corresponding to the trio data but with larger sample size than HapMap. Next strongest inflation is observed for the Sorbs, and finally, for a synthetic family study with a more extreme relatedness structure but with similar sample size as the Sorbs. Type I error increases rapidly with increasing inflation. However, for smaller significance levels, power increases with increasing inflation while the opposite holds for larger significance levels. When genomic control is applied, type I error is preserved while power decreases rapidly with increasing variance inflation. Stronger relatedness as well as higher heritability result in increased variance of the effect estimate of simple linear regression analysis. While type I error rates are generally inflated, the behaviour of power is more complex since power can be increased or reduced in dependence on relatedness and the heritability of the phenotype. Genomic control cannot be recommended to deal with inflation due to relatedness. Although it preserves type I error, the loss in power can be considerable. We provide a simple formula for estimating variance inflation given the relatedness structure and the heritability of a trait of interest. As a rule of thumb, variance inflation below 1.05 does not require correction and simple linear regression analysis is still appropriate.
NASA Astrophysics Data System (ADS)
Reis, D. S.; Stedinger, J. R.; Martins, E. S.
2005-10-01
This paper develops a Bayesian approach to analysis of a generalized least squares (GLS) regression model for regional analyses of hydrologic data. The new approach allows computation of the posterior distributions of the parameters and the model error variance using a quasi-analytic approach. Two regional skew estimation studies illustrate the value of the Bayesian GLS approach for regional statistical analysis of a shape parameter and demonstrate that regional skew models can be relatively precise with effective record lengths in excess of 60 years. With Bayesian GLS the marginal posterior distribution of the model error variance and the corresponding mean and variance of the parameters can be computed directly, thereby providing a simple but important extension of the regional GLS regression procedures popularized by Tasker and Stedinger (1989), which is sensitive to the likely values of the model error variance when it is small relative to the sampling error in the at-site estimator.
ERIC Educational Resources Information Center
Neel, John H.; Stallings, William M.
An influential statistics test recommends a Levene text for homogeneity of variance. A recent note suggests that Levene's test is upwardly biased for small samples. Another report shows inflated Alpha estimates and low power. Neither study utilized more than two sample sizes. This Monte Carlo study involved sampling from a normal population for…
A Method for Calculating the Mean Orbits of Meteor Streams
NASA Astrophysics Data System (ADS)
Voloshchuk, Yu. I.; Kashcheev, B. L.
An examination of the published catalogs of orbits of meteor streams and of a large number of works devoted to the selection of streams, their analysis and interpretation, showed that elements of stream orbits are calculated, as a rule, as arithmetical (sometimes, weighed) sample means. On the basis of these means, a search for parent bodies, a study of the evolution of swarms generating these streams, an analysis of one-dimensional and multidimensional distributions of these elements, etc., are performed. We show that systematic errors in the estimates of elements of the mean orbits are present in each of the catalogs. These errors are caused by the formal averaging of orbital elements over the sample, while ignoring the fact that they represent not only correlated, but dependent quantities, with nonlinear, in most cases, interrelations between them. Numerous examples are given of such inaccuracies, in particular, the cases where the "mean orbit of the stream" recorded by ground-based techniques does not cross the Earth's orbit. We suggest the computation algorithm, in which the averaging over the sample is carried out at the initial stage of the calculation of the mean orbit, and only for the variables required for subsequent calculations. After this, the known astrometric formulas are used to sequentially calculate all other parameters of the stream, considered now as a standard orbit. Variance analysis is used to estimate the errors in orbital elements of the streams, in the case that their orbits are obtained by averaging the orbital elements of meteoroids forming the stream, without taking into account their interdependence. The results obtained in this analysis indicate the behavior of systematic errors in the elements of orbits of meteor streams. As an example, the effect of the incorrect computation method on the distribution of elements of the stream orbits close to the orbits of asteroids of the Apollo, Aten, and Amor groups (AAA asteroids) is examined.
Improving the analysis of composite endpoints in rare disease trials.
McMenamin, Martina; Berglind, Anna; Wason, James M S
2018-05-22
Composite endpoints are recommended in rare diseases to increase power and/or to sufficiently capture complexity. Often, they are in the form of responder indices which contain a mixture of continuous and binary components. Analyses of these outcomes typically treat them as binary, thus only using the dichotomisations of continuous components. The augmented binary method offers a more efficient alternative and is therefore especially useful for rare diseases. Previous work has indicated the method may have poorer statistical properties when the sample size is small. Here we investigate small sample properties and implement small sample corrections. We re-sample from a previous trial with sample sizes varying from 30 to 80. We apply the standard binary and augmented binary methods and determine the power, type I error rate, coverage and average confidence interval width for each of the estimators. We implement Firth's adjustment for the binary component models and a small sample variance correction for the generalized estimating equations, applying the small sample adjusted methods to each sub-sample as before for comparison. For the log-odds treatment effect the power of the augmented binary method is 20-55% compared to 12-20% for the standard binary method. Both methods have approximately nominal type I error rates. The difference in response probabilities exhibit similar power but both unadjusted methods demonstrate type I error rates of 6-8%. The small sample corrected methods have approximately nominal type I error rates. On both scales, the reduction in average confidence interval width when using the adjusted augmented binary method is 17-18%. This is equivalent to requiring a 32% smaller sample size to achieve the same statistical power. The augmented binary method with small sample corrections provides a substantial improvement for rare disease trials using composite endpoints. We recommend the use of the method for the primary analysis in relevant rare disease trials. We emphasise that the method should be used alongside other efforts in improving the quality of evidence generated from rare disease trials rather than replace them.
Ziebart, Christina; Giangregorio, Lora M; Gibbs, Jenna C; Levine, Iris C; Tung, James; Laing, Andrew C
2017-06-14
A wide variety of accelerometer systems, with differing sensor characteristics, are used to detect impact loading during physical activities. The study examined the effects of system characteristics on measured peak impact loading during a variety of activities by comparing outputs from three separate accelerometer systems, and by assessing the influence of simulated reductions in operating range and sampling rate. Twelve healthy young adults performed seven tasks (vertical jump, box drop, heel drop, and bilateral single leg and lateral jumps) while simultaneously wearing three tri-axial accelerometers including a criterion standard laboratory-grade unit (Endevco 7267A) and two systems primarily used for activity-monitoring (ActiGraph GT3X+, GCDC X6-2mini). Peak acceleration (gmax) was compared across accelerometers, and errors resulting from down-sampling (from 640 to 100Hz) and range-limiting (to ±6g) the criterion standard output were characterized. The Actigraph activity-monitoring accelerometer underestimated gmax by an average of 30.2%; underestimation by the X6-2mini was not significant. Underestimation error was greater for tasks with greater impact magnitudes. gmax was underestimated when the criterion standard signal was down-sampled (by an average of 11%), range limited (by 11%), and by combined down-sampling and range-limiting (by 18%). These effects explained 89% of the variance in gmax error for the Actigraph system. This study illustrates that both the type and intensity of activity should be considered when selecting an accelerometer for characterizing impact events. In addition, caution may be warranted when comparing impact magnitudes from studies that use different accelerometers, and when comparing accelerometer outputs to osteogenic impact thresholds proposed in literature. Crown Copyright © 2017. Published by Elsevier Ltd. All rights reserved.
NASA Astrophysics Data System (ADS)
Gao, Jing; Burt, James E.
2017-12-01
This study investigates the usefulness of a per-pixel bias-variance error decomposition (BVD) for understanding and improving spatially-explicit data-driven models of continuous variables in environmental remote sensing (ERS). BVD is a model evaluation method originated from machine learning and have not been examined for ERS applications. Demonstrated with a showcase regression tree model mapping land imperviousness (0-100%) using Landsat images, our results showed that BVD can reveal sources of estimation errors, map how these sources vary across space, reveal the effects of various model characteristics on estimation accuracy, and enable in-depth comparison of different error metrics. Specifically, BVD bias maps can help analysts identify and delineate model spatial non-stationarity; BVD variance maps can indicate potential effects of ensemble methods (e.g. bagging), and inform efficient training sample allocation - training samples should capture the full complexity of the modeled process, and more samples should be allocated to regions with more complex underlying processes rather than regions covering larger areas. Through examining the relationships between model characteristics and their effects on estimation accuracy revealed by BVD for both absolute and squared errors (i.e. error is the absolute or the squared value of the difference between observation and estimate), we found that the two error metrics embody different diagnostic emphases, can lead to different conclusions about the same model, and may suggest different solutions for performance improvement. We emphasize BVD's strength in revealing the connection between model characteristics and estimation accuracy, as understanding this relationship empowers analysts to effectively steer performance through model adjustments.
Smoothed Spectra, Ogives, and Error Estimates for Atmospheric Turbulence Data
NASA Astrophysics Data System (ADS)
Dias, Nelson Luís
2018-01-01
A systematic evaluation is conducted of the smoothed spectrum, which is a spectral estimate obtained by averaging over a window of contiguous frequencies. The technique is extended to the ogive, as well as to the cross-spectrum. It is shown that, combined with existing variance estimates for the periodogram, the variance—and therefore the random error—associated with these estimates can be calculated in a straightforward way. The smoothed spectra and ogives are biased estimates; with simple power-law analytical models, correction procedures are devised, as well as a global constraint that enforces Parseval's identity. Several new results are thus obtained: (1) The analytical variance estimates compare well with the sample variance calculated for the Bartlett spectrum and the variance of the inertial subrange of the cospectrum is shown to be relatively much larger than that of the spectrum. (2) Ogives and spectra estimates with reduced bias are calculated. (3) The bias of the smoothed spectrum and ogive is shown to be negligible at the higher frequencies. (4) The ogives and spectra thus calculated have better frequency resolution than the Bartlett spectrum, with (5) gradually increasing variance and relative error towards the low frequencies. (6) Power-law identification and extraction of the rate of dissipation of turbulence kinetic energy are possible directly from the ogive. (7) The smoothed cross-spectrum is a valid inner product and therefore an acceptable candidate for coherence and spectral correlation coefficient estimation by means of the Cauchy-Schwarz inequality. The quadrature, phase function, coherence function and spectral correlation function obtained from the smoothed spectral estimates compare well with the classical ones derived from the Bartlett spectrum.
Inventory implications of using sampling variances in estimation of growth model coefficients
Albert R. Stage; William R. Wykoff
2000-01-01
Variables based on stand densities or stocking have sampling errors that depend on the relation of tree size to plot size and on the spatial structure of the population, ignoring the sampling errors of such variables, which include most measures of competition used in both distance-dependent and distance-independent growth models, can bias the predictions obtained from...
Global Sensitivity Analysis and Parameter Calibration for an Ecosystem Carbon Model
NASA Astrophysics Data System (ADS)
Safta, C.; Ricciuto, D. M.; Sargsyan, K.; Najm, H. N.; Debusschere, B.; Thornton, P. E.
2013-12-01
We present uncertainty quantification results for a process-based ecosystem carbon model. The model employs 18 parameters and is driven by meteorological data corresponding to years 1992-2006 at the Harvard Forest site. Daily Net Ecosystem Exchange (NEE) observations were available to calibrate the model parameters and test the performance of the model. Posterior distributions show good predictive capabilities for the calibrated model. A global sensitivity analysis was first performed to determine the important model parameters based on their contribution to the variance of NEE. We then proceed to calibrate the model parameters in a Bayesian framework. The daily discrepancies between measured and predicted NEE values were modeled as independent and identically distributed Gaussians with prescribed daily variance according to the recorded instrument error. All model parameters were assumed to have uninformative priors with bounds set according to expert opinion. The global sensitivity results show that the rate of leaf fall (LEAFALL) is responsible for approximately 25% of the total variance in the average NEE for 1992-2005. A set of 4 other parameters, Nitrogen use efficiency (NUE), base rate for maintenance respiration (BR_MR), growth respiration fraction (RG_FRAC), and allocation to plant stem pool (ASTEM) contribute between 5% and 12% to the variance in average NEE, while the rest of the parameters have smaller contributions. The posterior distributions, sampled with a Markov Chain Monte Carlo algorithm, exhibit significant correlations between model parameters. However LEAFALL, the most important parameter for the average NEE, is not informed by the observational data, while less important parameters show significant updates between their prior and posterior densities. The Fisher information matrix values, indicating which parameters are most informed by the experimental observations, are examined to augment the comparison between the calibration and global sensitivity analysis results.
Out-of-plane ultrasonic velocity measurement
Hall, M.S.; Brodeur, P.H.; Jackson, T.G.
1998-07-14
A method for improving the accuracy of measuring the velocity and time of flight of ultrasonic signals through moving web-like materials such as paper, paperboard and the like, includes a pair of ultrasonic transducers disposed on opposing sides of a moving web-like material. In order to provide acoustical coupling between the transducers and the web-like material, the transducers are disposed in fluid-filled wheels. Errors due to variances in the wheel thicknesses about their circumference which can affect time of flight measurements and ultimately the mechanical property being tested are compensated by averaging the ultrasonic signals for a predetermined number of revolutions. The invention further includes a method for compensating for errors resulting from the digitization of the ultrasonic signals. More particularly, the invention includes a method for eliminating errors known as trigger jitter inherent with digitizing oscilloscopes used to digitize the signals for manipulation by a digital computer. In particular, rather than cross-correlate ultrasonic signals taken during different sample periods as is known in the art in order to determine the time of flight of the ultrasonic signal through the moving web, a pulse echo box is provided to enable cross-correlation of predetermined transmitted ultrasonic signals with predetermined reflected ultrasonic or echo signals during the sample period. By cross-correlating ultrasonic signals in the same sample period, the error associated with trigger jitter is eliminated. 20 figs.
Out-of-plane ultrasonic velocity measurement
Hall, Maclin S.; Brodeur, Pierre H.; Jackson, Theodore G.
1998-01-01
A method for improving the accuracy of measuring the velocity and time of flight of ultrasonic signals through moving web-like materials such as paper, paperboard and the like, includes a pair of ultrasonic transducers disposed on opposing sides of a moving web-like material. In order to provide acoustical coupling between the transducers and the web-like material, the transducers are disposed in fluid-filled wheels. Errors due to variances in the wheel thicknesses about their circumference which can affect time of flight measurements and ultimately the mechanical property being tested are compensated by averaging the ultrasonic signals for a predetermined number of revolutions. The invention further includes a method for compensating for errors resulting from the digitization of the ultrasonic signals. More particularly, the invention includes a method for eliminating errors known as trigger jitter inherent with digitizing oscilloscopes used to digitize the signals for manipulation by a digital computer. In particular, rather than cross-correlate ultrasonic signals taken during different sample periods as is known in the art in order to determine the time of flight of the ultrasonic signal through the moving web, a pulse echo box is provided to enable cross-correlation of predetermined transmitted ultrasonic signals with predetermined reflected ultrasonic or echo signals during the sample period. By cross-correlating ultrasonic signals in the same sample period, the error associated with trigger jitter is eliminated.
Distribution of kriging errors, the implications and how to communicate them
NASA Astrophysics Data System (ADS)
Li, Hong Yi; Milne, Alice; Webster, Richard
2016-04-01
Kriging in one form or another has become perhaps the most popular method for spatial prediction in environmental science. Each prediction is unbiased and of minimum variance, which itself is estimated. The kriging variances depend on the mathematical model chosen to describe the spatial variation; different models, however plausible, give rise to different minimized variances. Practitioners often compare models by so-called cross-validation before finally choosing the most appropriate for their kriging. One proceeds as follows. One removes a unit (a sampling point) from the whole set, kriges the value there and compares the kriged value with the value observed to obtain the deviation or error. One repeats the process for each and every point in turn and for all plausible models. One then computes the mean errors (MEs) and the mean of the squared errors (MSEs). Ideally a squared error should equal the corresponding kriging variance (σK2), and so one is advised to choose the model for which on average the squared errors most nearly equal the kriging variances, i.e. the ratio MSDR = MSE/σK2 ≈ 1. Maximum likelihood estimation of models almost guarantees that the MSDR equals 1, and so the kriging variances are unbiased predictors of the squared error across the region. The method is based on the assumption that the errors have a normal distribution. The squared deviation ratio (SDR) should therefore be distributed as χ2 with one degree of freedom with a median of 0.455. We have found that often the median of the SDR (MedSDR) is less, in some instances much less, than 0.455 even though the mean of the SDR is close to 1. It seems that in these cases the distributions of the errors are leptokurtic, i.e. they have an excess of predictions close to the true values, excesses near the extremes and a dearth of predictions in between. In these cases the kriging variances are poor measures of the uncertainty at individual sites. The uncertainty is typically under-estimated for the extreme observations and compensated for by over estimating for other observations. Statisticians must tell users when they present maps of predictions. We illustrate the situation with results from mapping salinity in land reclaimed from the Yangtze delta in the Gulf of Hangzhou, China. There the apparent electrical conductivity (ECa) of the topsoil was measured at 525 points in a field of 2.3 ha. The marginal distribution of the observations was strongly positively skewed, and so the observed ECas were transformed to their logarithms to give an approximately symmetric distribution. That distribution was strongly platykurtic with short tails and no evident outliers. The logarithms were analysed as a mixed model of quadratic drift plus correlated random residuals with a spherical variogram. The kriged predictions that deviated from their true values with an MSDR of 0.993, but with a medSDR=0.324. The coefficient of kurtosis of the deviations was 1.45, i.e. substantially larger than 0 for a normal distribution. The reasons for this behaviour are being sought. The most likely explanation is that there are spatial outliers, i.e. points at which the observed values that differ markedly from those at their their closest neighbours.
Distribution of kriging errors, the implications and how to communicate them
NASA Astrophysics Data System (ADS)
Li, HongYi; Milne, Alice; Webster, Richard
2015-04-01
Kriging in one form or another has become perhaps the most popular method for spatial prediction in environmental science. Each prediction is unbiased and of minimum variance, which itself is estimated. The kriging variances depend on the mathematical model chosen to describe the spatial variation; different models, however plausible, give rise to different minimized variances. Practitioners often compare models by so-called cross-validation before finally choosing the most appropriate for their kriging. One proceeds as follows. One removes a unit (a sampling point) from the whole set, kriges the value there and compares the kriged value with the value observed to obtain the deviation or error. One repeats the process for each and every point in turn and for all plausible models. One then computes the mean errors (MEs) and the mean of the squared errors (MSEs). Ideally a squared error should equal the corresponding kriging variance (σ_K^2), and so one is advised to choose the model for which on average the squared errors most nearly equal the kriging variances, i.e. the ratio MSDR=MSE/ σ_K2 ≈1. Maximum likelihood estimation of models almost guarantees that the MSDR equals 1, and so the kriging variances are unbiased predictors of the squared error across the region. The method is based on the assumption that the errors have a normal distribution. The squared deviation ratio (SDR) should therefore be distributed as χ2 with one degree of freedom with a median of 0.455. We have found that often the median of the SDR (MedSDR) is less, in some instances much less, than 0.455 even though the mean of the SDR is close to 1. It seems that in these cases the distributions of the errors are leptokurtic, i.e. they have an excess of predictions close to the true values, excesses near the extremes and a dearth of predictions in between. In these cases the kriging variances are poor measures of the uncertainty at individual sites. The uncertainty is typically under-estimated for the extreme observations and compensated for by over estimating for other observations. Statisticians must tell users when they present maps of predictions. We illustrate the situation with results from mapping salinity in land reclaimed from the Yangtze delta in the Gulf of Hangzhou, China. There the apparent electrical conductivity (EC_a) of the topsoil was measured at 525 points in a field of 2.3~ha. The marginal distribution of the observations was strongly positively skewed, and so the observed EC_as were transformed to their logarithms to give an approximately symmetric distribution. That distribution was strongly platykurtic with short tails and no evident outliers. The logarithms were analysed as a mixed model of quadratic drift plus correlated random residuals with a spherical variogram. The kriged predictions that deviated from their true values with an MSDR of 0.993, but with a medSDR=0.324. The coefficient of kurtosis of the deviations was 1.45, i.e. substantially larger than 0 for a normal distribution. The reasons for this behaviour are being sought. The most likely explanation is that there are spatial outliers, i.e. points at which the observed values that differ markedly from those at their their closest neighbours.
Johnson, Jacqueline L; Kreidler, Sarah M; Catellier, Diane J; Murray, David M; Muller, Keith E; Glueck, Deborah H
2015-11-30
We used theoretical and simulation-based approaches to study Type I error rates for one-stage and two-stage analytic methods for cluster-randomized designs. The one-stage approach uses the observed data as outcomes and accounts for within-cluster correlation using a general linear mixed model. The two-stage model uses the cluster specific means as the outcomes in a general linear univariate model. We demonstrate analytically that both one-stage and two-stage models achieve exact Type I error rates when cluster sizes are equal. With unbalanced data, an exact size α test does not exist, and Type I error inflation may occur. Via simulation, we compare the Type I error rates for four one-stage and six two-stage hypothesis testing approaches for unbalanced data. With unbalanced data, the two-stage model, weighted by the inverse of the estimated theoretical variance of the cluster means, and with variance constrained to be positive, provided the best Type I error control for studies having at least six clusters per arm. The one-stage model with Kenward-Roger degrees of freedom and unconstrained variance performed well for studies having at least 14 clusters per arm. The popular analytic method of using a one-stage model with denominator degrees of freedom appropriate for balanced data performed poorly for small sample sizes and low intracluster correlation. Because small sample sizes and low intracluster correlation are common features of cluster-randomized trials, the Kenward-Roger method is the preferred one-stage approach. Copyright © 2015 John Wiley & Sons, Ltd.
NASA Astrophysics Data System (ADS)
Wang, Ting; Xiang, Jie; Fei, Jianfang; Wang, Yi; Liu, Chunxia; Li, Yuanxiang
2017-12-01
This paper presents an evaluation of the observational impacts on blended sea surface winds from a two-dimensional variational data assimilation (2D-Var) scheme. We begin by briefly introducing the analysis sensitivity with respect to observations in variational data assimilation systems and its relationship with the degrees of freedom for signal (DFS), and then the DFS concept is applied to the 2D-Var sea surface wind blending scheme. Two methods, a priori and a posteriori, are used to estimate the DFS of the zonal ( u) and meridional ( v) components of winds in the 2D-Var blending scheme. The a posteriori method can obtain almost the same results as the a priori method. Because only by-products of the blending scheme are used for the a posteriori method, the computation time is reduced significantly. The magnitude of the DFS is critically related to the observational and background error statistics. Changing the observational and background error variances can affect the DFS value. Because the observation error variances are assumed to be uniform, the observational influence at each observational location is related to the background error variance, and the observations located at the place where there are larger background error variances have larger influences. The average observational influence of u and v with respect to the analysis is about 40%, implying that the background influence with respect to the analysis is about 60%.
Accounting for Sampling Error in Genetic Eigenvalues Using Random Matrix Theory.
Sztepanacz, Jacqueline L; Blows, Mark W
2017-07-01
The distribution of genetic variance in multivariate phenotypes is characterized by the empirical spectral distribution of the eigenvalues of the genetic covariance matrix. Empirical estimates of genetic eigenvalues from random effects linear models are known to be overdispersed by sampling error, where large eigenvalues are biased upward, and small eigenvalues are biased downward. The overdispersion of the leading eigenvalues of sample covariance matrices have been demonstrated to conform to the Tracy-Widom (TW) distribution. Here we show that genetic eigenvalues estimated using restricted maximum likelihood (REML) in a multivariate random effects model with an unconstrained genetic covariance structure will also conform to the TW distribution after empirical scaling and centering. However, where estimation procedures using either REML or MCMC impose boundary constraints, the resulting genetic eigenvalues tend not be TW distributed. We show how using confidence intervals from sampling distributions of genetic eigenvalues without reference to the TW distribution is insufficient protection against mistaking sampling error as genetic variance, particularly when eigenvalues are small. By scaling such sampling distributions to the appropriate TW distribution, the critical value of the TW statistic can be used to determine if the magnitude of a genetic eigenvalue exceeds the sampling error for each eigenvalue in the spectral distribution of a given genetic covariance matrix. Copyright © 2017 by the Genetics Society of America.
The Statistical Power of Planned Comparisons.
ERIC Educational Resources Information Center
Benton, Roberta L.
Basic principles underlying statistical power are examined; and issues pertaining to effect size, sample size, error variance, and significance level are highlighted via the use of specific hypothetical examples. Analysis of variance (ANOVA) and related methods remain popular, although other procedures sometimes have more statistical power against…
Increasing point-count duration increases standard error
Smith, W.P.; Twedt, D.J.; Hamel, P.B.; Ford, R.P.; Wiedenfeld, D.A.; Cooper, R.J.
1998-01-01
We examined data from point counts of varying duration in bottomland forests of west Tennessee and the Mississippi Alluvial Valley to determine if counting interval influenced sampling efficiency. Estimates of standard error increased as point count duration increased both for cumulative number of individuals and species in both locations. Although point counts appear to yield data with standard errors proportional to means, a square root transformation of the data may stabilize the variance. Using long (>10 min) point counts may reduce sample size and increase sampling error, both of which diminish statistical power and thereby the ability to detect meaningful changes in avian populations.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Wang, Yi-Xin; Zeng, Qiang; Wang, Le
Urinary haloacetic acids (HAAs), such as dichloroacetic acid (DCAA) and trichloroacetic acid (TCAA), have been suggested as potential biomarkers of exposure to drinking water disinfection byproducts (DBPs). However, variable exposure to and the short elimination half-lives of these biomarkers can result in considerable variability in urinary measurements, leading to exposure misclassification. Here we examined the variability of DCAA and TCAA levels in the urine among eleven men who provided urine samples on 8 days over 3 months. The urinary concentrations of DCAA and TCAA were measured by gas chromatography coupled with electron capture detection. We calculated the intraclass correlation coefficientsmore » (ICCs) to characterize the within-person and between-person variances and computed the sensitivity and specificity to assess how well single or multiple urine collections accurately determined personal 3-month average DCAA and TCAA levels. The within-person variance was much higher than the between-person variance for all three sample types (spot, first morning, and 24-h urine samples) for DCAA (ICC=0.08–0.37) and TCAA (ICC=0.09–0.23), regardless of the sampling interval. A single-spot urinary sample predicted high (top 33%) 3-month average DCAA and TCAA levels with high specificity (0.79 and 0.78, respectively) but relatively low sensitivity (0.47 and 0.50, respectively). Collecting two or three urine samples from each participant improved the classification. The poor reproducibility of the measured urinary DCAA and TCAA concentrations indicate that a single measurement may not accurately reflect individual long-term exposure. Collection of multiple urine samples from one person is an option for reducing exposure classification errors in studies exploring the effects of DBP exposure on reproductive health. - Highlights: • We evaluated the variability of DCAA and TCAA levels in the urine among men. • Urinary DCAA and TCAA levels varied greatly over a 3-month period. • Single measurement may not accurately reflect personal long-term exposure levels. • Collecting multiple samples from one person improved the exposure classification.« less
Mixed emotions: Sensitivity to facial variance in a crowd of faces.
Haberman, Jason; Lee, Pegan; Whitney, David
2015-01-01
The visual system automatically represents summary information from crowds of faces, such as the average expression. This is a useful heuristic insofar as it provides critical information about the state of the world, not simply information about the state of one individual. However, the average alone is not sufficient for making decisions about how to respond to a crowd. The variance or heterogeneity of the crowd--the mixture of emotions--conveys information about the reliability of the average, essential for determining whether the average can be trusted. Despite its importance, the representation of variance within a crowd of faces has yet to be examined. This is addressed here in three experiments. In the first experiment, observers viewed a sample set of faces that varied in emotion, and then adjusted a subsequent set to match the variance of the sample set. To isolate variance as the summary statistic of interest, the average emotion of both sets was random. Results suggested that observers had information regarding crowd variance. The second experiment verified that this was indeed a uniquely high-level phenomenon, as observers were unable to derive the variance of an inverted set of faces as precisely as an upright set of faces. The third experiment replicated and extended the first two experiments using method-of-constant-stimuli. Together, these results show that the visual system is sensitive to emergent information about the emotional heterogeneity, or ambivalence, in crowds of faces.
Brandmaier, Andreas M.; von Oertzen, Timo; Ghisletta, Paolo; Lindenberger, Ulman; Hertzog, Christopher
2018-01-01
Latent Growth Curve Models (LGCM) have become a standard technique to model change over time. Prediction and explanation of inter-individual differences in change are major goals in lifespan research. The major determinants of statistical power to detect individual differences in change are the magnitude of true inter-individual differences in linear change (LGCM slope variance), design precision, alpha level, and sample size. Here, we show that design precision can be expressed as the inverse of effective error. Effective error is determined by instrument reliability and the temporal arrangement of measurement occasions. However, it also depends on another central LGCM component, the variance of the latent intercept and its covariance with the latent slope. We derive a new reliability index for LGCM slope variance—effective curve reliability (ECR)—by scaling slope variance against effective error. ECR is interpretable as a standardized effect size index. We demonstrate how effective error, ECR, and statistical power for a likelihood ratio test of zero slope variance formally relate to each other and how they function as indices of statistical power. We also provide a computational approach to derive ECR for arbitrary intercept-slope covariance. With practical use cases, we argue for the complementary utility of the proposed indices of a study's sensitivity to detect slope variance when making a priori longitudinal design decisions or communicating study designs. PMID:29755377
Estimating means and variances: The comparative efficiency of composite and grab samples.
Brumelle, S; Nemetz, P; Casey, D
1984-03-01
This paper compares the efficiencies of two sampling techniques for estimating a population mean and variance. One procedure, called grab sampling, consists of collecting and analyzing one sample per period. The second procedure, called composite sampling, collectsn samples per period which are then pooled and analyzed as a single sample. We review the well known fact that composite sampling provides a superior estimate of the mean. However, it is somewhat surprising that composite sampling does not always generate a more efficient estimate of the variance. For populations with platykurtic distributions, grab sampling gives a more efficient estimate of the variance, whereas composite sampling is better for leptokurtic distributions. These conditions on kurtosis can be related to peakedness and skewness. For example, a necessary condition for composite sampling to provide a more efficient estimate of the variance is that the population density function evaluated at the mean (i.e.f(μ)) be greater than[Formula: see text]. If[Formula: see text], then a grab sample is more efficient. In spite of this result, however, composite sampling does provide a smaller estimate of standard error than does grab sampling in the context of estimating population means.
A note on the kappa statistic for clustered dichotomous data.
Zhou, Ming; Yang, Zhao
2014-06-30
The kappa statistic is widely used to assess the agreement between two raters. Motivated by a simulation-based cluster bootstrap method to calculate the variance of the kappa statistic for clustered physician-patients dichotomous data, we investigate its special correlation structure and develop a new simple and efficient data generation algorithm. For the clustered physician-patients dichotomous data, based on the delta method and its special covariance structure, we propose a semi-parametric variance estimator for the kappa statistic. An extensive Monte Carlo simulation study is performed to evaluate the performance of the new proposal and five existing methods with respect to the empirical coverage probability, root-mean-square error, and average width of the 95% confidence interval for the kappa statistic. The variance estimator ignoring the dependence within a cluster is generally inappropriate, and the variance estimators from the new proposal, bootstrap-based methods, and the sampling-based delta method perform reasonably well for at least a moderately large number of clusters (e.g., the number of clusters K ⩾50). The new proposal and sampling-based delta method provide convenient tools for efficient computations and non-simulation-based alternatives to the existing bootstrap-based methods. Moreover, the new proposal has acceptable performance even when the number of clusters is as small as K = 25. To illustrate the practical application of all the methods, one psychiatric research data and two simulated clustered physician-patients dichotomous data are analyzed. Copyright © 2014 John Wiley & Sons, Ltd.
On-line estimation of error covariance parameters for atmospheric data assimilation
NASA Technical Reports Server (NTRS)
Dee, Dick P.
1995-01-01
A simple scheme is presented for on-line estimation of covariance parameters in statistical data assimilation systems. The scheme is based on a maximum-likelihood approach in which estimates are produced on the basis of a single batch of simultaneous observations. Simple-sample covariance estimation is reasonable as long as the number of available observations exceeds the number of tunable parameters by two or three orders of magnitude. Not much is known at present about model error associated with actual forecast systems. Our scheme can be used to estimate some important statistical model error parameters such as regionally averaged variances or characteristic correlation length scales. The advantage of the single-sample approach is that it does not rely on any assumptions about the temporal behavior of the covariance parameters: time-dependent parameter estimates can be continuously adjusted on the basis of current observations. This is of practical importance since it is likely to be the case that both model error and observation error strongly depend on the actual state of the atmosphere. The single-sample estimation scheme can be incorporated into any four-dimensional statistical data assimilation system that involves explicit calculation of forecast error covariances, including optimal interpolation (OI) and the simplified Kalman filter (SKF). The computational cost of the scheme is high but not prohibitive; on-line estimation of one or two covariance parameters in each analysis box of an operational bozed-OI system is currently feasible. A number of numerical experiments performed with an adaptive SKF and an adaptive version of OI, using a linear two-dimensional shallow-water model and artificially generated model error are described. The performance of the nonadaptive versions of these methods turns out to depend rather strongly on correct specification of model error parameters. These parameters are estimated under a variety of conditions, including uniformly distributed model error and time-dependent model error statistics.
NASA Technical Reports Server (NTRS)
Nese, Jon M.; Dutton, John A.
1993-01-01
The predictability of the weather and climatic states of a low-order moist general circulation model is quantified using a dynamic systems approach, and the effect of incorporating a simple oceanic circulation on predictability is evaluated. The predictability and the structure of the model attractors are compared using Liapunov exponents, local divergence rates, and the correlation and Liapunov dimensions. It was found that the activation of oceanic circulation increases the average error doubling time of the atmosphere and the coupled ocean-atmosphere system by 10 percent and decreases the variance of the largest local divergence rate by 20 percent. When an oceanic circulation develops, the average predictability of annually averaged states is improved by 25 percent and the variance of the largest local divergence rate decreases by 25 percent.
Estimation of genetic parameters for milk yield in Murrah buffaloes by Bayesian inference.
Breda, F C; Albuquerque, L G; Euclydes, R F; Bignardi, A B; Baldi, F; Torres, R A; Barbosa, L; Tonhati, H
2010-02-01
Random regression models were used to estimate genetic parameters for test-day milk yield in Murrah buffaloes using Bayesian inference. Data comprised 17,935 test-day milk records from 1,433 buffaloes. Twelve models were tested using different combinations of third-, fourth-, fifth-, sixth-, and seventh-order orthogonal polynomials of weeks of lactation for additive genetic and permanent environmental effects. All models included the fixed effects of contemporary group, number of daily milkings and age of cow at calving as covariate (linear and quadratic effect). In addition, residual variances were considered to be heterogeneous with 6 classes of variance. Models were selected based on the residual mean square error, weighted average of residual variance estimates, and estimates of variance components, heritabilities, correlations, eigenvalues, and eigenfunctions. Results indicated that changes in the order of fit for additive genetic and permanent environmental random effects influenced the estimation of genetic parameters. Heritability estimates ranged from 0.19 to 0.31. Genetic correlation estimates were close to unity between adjacent test-day records, but decreased gradually as the interval between test-days increased. Results from mean squared error and weighted averages of residual variance estimates suggested that a model considering sixth- and seventh-order Legendre polynomials for additive and permanent environmental effects, respectively, and 6 classes for residual variances, provided the best fit. Nevertheless, this model presented the largest degree of complexity. A more parsimonious model, with fourth- and sixth-order polynomials, respectively, for these same effects, yielded very similar genetic parameter estimates. Therefore, this last model is recommended for routine applications. Copyright 2010 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.
Location tests for biomarker studies: a comparison using simulations for the two-sample case.
Scheinhardt, M O; Ziegler, A
2013-01-01
Gene, protein, or metabolite expression levels are often non-normally distributed, heavy tailed and contain outliers. Standard statistical approaches may fail as location tests in this situation. In three Monte-Carlo simulation studies, we aimed at comparing the type I error levels and empirical power of standard location tests and three adaptive tests [O'Gorman, Can J Stat 1997; 25: 269 -279; Keselman et al., Brit J Math Stat Psychol 2007; 60: 267- 293; Szymczak et al., Stat Med 2013; 32: 524 - 537] for a wide range of distributions. We simulated two-sample scenarios using the g-and-k-distribution family to systematically vary tail length and skewness with identical and varying variability between groups. All tests kept the type I error level when groups did not vary in their variability. The standard non-parametric U-test performed well in all simulated scenarios. It was outperformed by the two non-parametric adaptive methods in case of heavy tails or large skewness. Most tests did not keep the type I error level for skewed data in the case of heterogeneous variances. The standard U-test was a powerful and robust location test for most of the simulated scenarios except for very heavy tailed or heavy skewed data, and it is thus to be recommended except for these cases. The non-parametric adaptive tests were powerful for both normal and non-normal distributions under sample variance homogeneity. But when sample variances differed, they did not keep the type I error level. The parametric adaptive test lacks power for skewed and heavy tailed distributions.
SU-E-T-192: FMEA Severity Scores - Do We Really Know?
DOE Office of Scientific and Technical Information (OSTI.GOV)
Tonigan, J; Johnson, J; Kry, S
2014-06-01
Purpose: Failure modes and effects analysis (FMEA) is a subjective risk mitigation technique that has not been applied to physics-specific quality management practices. There is a need for quantitative FMEA data as called for in the literature. This work focuses specifically on quantifying FMEA severity scores for physics components of IMRT delivery and comparing to subjective scores. Methods: Eleven physical failure modes (FMs) for head and neck IMRT dose calculation and delivery are examined near commonly accepted tolerance criteria levels. Phantom treatment planning studies and dosimetry measurements (requiring decommissioning in several cases) are performed to determine the magnitude of dosemore » delivery errors for the FMs (i.e., severity of the FM). Resultant quantitative severity scores are compared to FMEA scores obtained through an international survey and focus group studies. Results: Physical measurements for six FMs have resulted in significant PTV dose errors up to 4.3% as well as close to 1 mm significant distance-to-agreement error between PTV and OAR. Of the 129 survey responses, the vast majority of the responders used Varian machines with Pinnacle and Eclipse planning systems. The average years of experience was 17, yet familiarity with FMEA less than expected. Survey reports perception of dose delivery error magnitude varies widely, in some cases 50% difference in dose delivery error expected amongst respondents. Substantial variance is also seen for all FMs in occurrence, detectability, and severity scores assigned with average variance values of 5.5, 4.6, and 2.2, respectively. Survey shows for MLC positional FM(2mm) average of 7.6% dose error expected (range 0–50%) compared to 2% error seen in measurement. Analysis of ranking in survey, treatment planning studies, and quantitative value comparison will be presented. Conclusion: Resultant quantitative severity scores will expand the utility of FMEA for radiotherapy and verify accuracy of FMEA results compared to highly variable subjective scores.« less
Berger, Philip; Messner, Michael J; Crosby, Jake; Vacs Renwick, Deborah; Heinrich, Austin
2018-05-01
Spore reduction can be used as a surrogate measure of Cryptosporidium natural filtration efficiency. Estimates of log10 (log) reduction were derived from spore measurements in paired surface and well water samples in Casper Wyoming and Kearney Nebraska. We found that these data were suitable for testing the hypothesis (H 0 ) that the average reduction at each site was 2 log or less, using a one-sided Student's t-test. After establishing data quality objectives for the test (expressed as tolerable Type I and Type II error rates), we evaluated the test's performance as a function of the (a) true log reduction, (b) number of paired samples assayed and (c) variance of observed log reductions. We found that 36 paired spore samples are sufficient to achieve the objectives over a wide range of variance, including the variances observed in the two data sets. We also explored the feasibility of using smaller numbers of paired spore samples to supplement bioparticle counts for screening purposes in alluvial aquifers, to differentiate wells with large volume surface water induced recharge from wells with negligible surface water induced recharge. With key assumptions, we propose a normal statistical test of the same hypothesis (H 0 ), but with different performance objectives. As few as six paired spore samples appear adequate as a screening metric to supplement bioparticle counts to differentiate wells in alluvial aquifers with large volume surface water induced recharge. For the case when all available information (including failure to reject H 0 based on the limited paired spore data) leads to the conclusion that wells have large surface water induced recharge, we recommend further evaluation using additional paired biweekly spore samples. Published by Elsevier GmbH.
NASA Astrophysics Data System (ADS)
Yang, Yanqiu; Yu, Lin; Zhang, Yixin
2017-04-01
A model of the average capacity of optical wireless communication link with pointing errors for the ground-to-train of the curved track is established based on the non-Kolmogorov. By adopting the gamma-gamma distribution model, we derive the average capacity expression for this channel. The numerical analysis reveals that heavier fog reduces the average capacity of link. The strength of atmospheric turbulence, the variance of pointing errors, and the covered track length need to be reduced for the larger average capacity of link. The normalized beamwidth and the average signal-to-noise ratio (SNR) of the turbulence-free link need to be increased. We can increase the transmit aperture to expand the beamwidth and enhance the signal intensity, thereby decreasing the impact of the beam wander accordingly. As the system adopting the automatic tracking of beam at the receiver positioned on the roof of the train, for eliminating the pointing errors caused by beam wander and train vibration, the equivalent average capacity of the channel will achieve a maximum value. The impact of the non-Kolmogorov spectral index's variation on the average capacity of link can be ignored.
NASA Astrophysics Data System (ADS)
Rexer, Moritz; Hirt, Christian
2015-09-01
Classical degree variance models (such as Kaula's rule or the Tscherning-Rapp model) often rely on low-resolution gravity data and so are subject to extrapolation when used to describe the decay of the gravity field at short spatial scales. This paper presents a new degree variance model based on the recently published GGMplus near-global land areas 220 m resolution gravity maps (Geophys Res Lett 40(16):4279-4283, 2013). We investigate and use a 2D-DFT (discrete Fourier transform) approach to transform GGMplus gravity grids into degree variances. The method is described in detail and its approximation errors are studied using closed-loop experiments. Focus is placed on tiling, azimuth averaging, and windowing effects in the 2D-DFT method and on analytical fitting of degree variances. Approximation errors of the 2D-DFT procedure on the (spherical harmonic) degree variance are found to be at the 10-20 % level. The importance of the reference surface (sphere, ellipsoid or topography) of the gravity data for correct interpretation of degree variance spectra is highlighted. The effect of the underlying mass arrangement (spherical or ellipsoidal approximation) on the degree variances is found to be crucial at short spatial scales. A rule-of-thumb for transformation of spectra between spherical and ellipsoidal approximation is derived. Application of the 2D-DFT on GGMplus gravity maps yields a new degree variance model to degree 90,000. The model is supported by GRACE, GOCE, EGM2008 and forward-modelled gravity at 3 billion land points over all land areas within the SRTM data coverage and provides gravity signal variances at the surface of the topography. The model yields omission errors of 9 mGal for gravity (1.5 cm for geoid effects) at scales of 10 km, 4 mGal (1 mm) at 2-km scales, and 2 mGal (0.2 mm) at 1-km scales.
Willem W.S. van Hees
2002-01-01
Comparisons of estimated standard error for a ratio-of-means (ROM) estimator are presented for forest resource inventories conducted in southeast Alaska between 1995 and 2000. Estimated standard errors for the ROM were generated by using a traditional variance estimator and also approximated by bootstrap methods. Estimates of standard error generated by both...
NASA Astrophysics Data System (ADS)
Söderberg, Per G.; Malmberg, Filip; Sandberg-Melin, Camilla
2016-03-01
The present study aimed to analyze the clinical usefulness of the thinnest cross section of the nerve fibers in the optic nerve head averaged over the circumference of the optic nerve head. 3D volumes of the optic nerve head of the same eye was captured at two different visits spaced in time by 1-4 weeks, in 13 subjects diagnosed with early to moderate glaucoma. At each visit 3 volumes containing the optic nerve head were captured independently with a Topcon OCT- 2000 system. In each volume, the average shortest distance between the inner surface of the retina and the central limit of the pigment epithelium around the optic nerve head circumference, PIMD-Average [02π], was determined semiautomatically. The measurements were analyzed with an analysis of variance for estimation of the variance components for subjects, visits, volumes and semi-automatic measurements of PIMD-Average [0;2π]. It was found that the variance for subjects was on the order of five times the variance for visits, and the variance for visits was on the order of 5 times higher than the variance for volumes. The variance for semi-automatic measurements of PIMD-Average [02π] was 3 orders of magnitude lower than the variance for volumes. A 95 % confidence interval for mean PIMD-Average [02π] was estimated to 1.00 +/-0.13 mm (D.f. = 12). The variance estimates indicate that PIMD-Average [02π] is not suitable for comparison between a onetime estimate in a subject and a population reference interval. Cross-sectional independent group comparisons of PIMD-Average [02π] averaged over subjects will require inconveniently large sample sizes. However, cross-sectional independent group comparison of averages of within subject difference between baseline and follow-up can be made with reasonable sample sizes. Assuming a loss rate of 0.1 PIMD-Average [02π] per year and 4 visits per year it was found that approximately 18 months follow up is required before a significant change of PIMDAverage [02π] can be observed with a power of 0.8. This is shorter than what has been observed both for HRT measurements and automated perimetry measurements with a similar observation rate. It is concluded that PIMDAverage [02π] has the potential to detect deterioration of glaucoma quicker than currently available primary diagnostic instruments. To increase the efficiency of PIMD-Average [02π] further, the variation among visits within subject has to be reduced.
Inference from clustering with application to gene-expression microarrays.
Dougherty, Edward R; Barrera, Junior; Brun, Marcel; Kim, Seungchan; Cesar, Roberto M; Chen, Yidong; Bittner, Michael; Trent, Jeffrey M
2002-01-01
There are many algorithms to cluster sample data points based on nearness or a similarity measure. Often the implication is that points in different clusters come from different underlying classes, whereas those in the same cluster come from the same class. Stochastically, the underlying classes represent different random processes. The inference is that clusters represent a partition of the sample points according to which process they belong. This paper discusses a model-based clustering toolbox that evaluates cluster accuracy. Each random process is modeled as its mean plus independent noise, sample points are generated, the points are clustered, and the clustering error is the number of points clustered incorrectly according to the generating random processes. Various clustering algorithms are evaluated based on process variance and the key issue of the rate at which algorithmic performance improves with increasing numbers of experimental replications. The model means can be selected by hand to test the separability of expected types of biological expression patterns. Alternatively, the model can be seeded by real data to test the expected precision of that output or the extent of improvement in precision that replication could provide. In the latter case, a clustering algorithm is used to form clusters, and the model is seeded with the means and variances of these clusters. Other algorithms are then tested relative to the seeding algorithm. Results are averaged over various seeds. Output includes error tables and graphs, confusion matrices, principal-component plots, and validation measures. Five algorithms are studied in detail: K-means, fuzzy C-means, self-organizing maps, hierarchical Euclidean-distance-based and correlation-based clustering. The toolbox is applied to gene-expression clustering based on cDNA microarrays using real data. Expression profile graphics are generated and error analysis is displayed within the context of these profile graphics. A large amount of generated output is available over the web.
Bartz, Daniel; Hatrick, Kerr; Hesse, Christian W; Müller, Klaus-Robert; Lemm, Steven
2013-01-01
Robust and reliable covariance estimates play a decisive role in financial and many other applications. An important class of estimators is based on factor models. Here, we show by extensive Monte Carlo simulations that covariance matrices derived from the statistical Factor Analysis model exhibit a systematic error, which is similar to the well-known systematic error of the spectrum of the sample covariance matrix. Moreover, we introduce the Directional Variance Adjustment (DVA) algorithm, which diminishes the systematic error. In a thorough empirical study for the US, European, and Hong Kong stock market we show that our proposed method leads to improved portfolio allocation.
Bartz, Daniel; Hatrick, Kerr; Hesse, Christian W.; Müller, Klaus-Robert; Lemm, Steven
2013-01-01
Robust and reliable covariance estimates play a decisive role in financial and many other applications. An important class of estimators is based on factor models. Here, we show by extensive Monte Carlo simulations that covariance matrices derived from the statistical Factor Analysis model exhibit a systematic error, which is similar to the well-known systematic error of the spectrum of the sample covariance matrix. Moreover, we introduce the Directional Variance Adjustment (DVA) algorithm, which diminishes the systematic error. In a thorough empirical study for the US, European, and Hong Kong stock market we show that our proposed method leads to improved portfolio allocation. PMID:23844016
Scott, JoAnna M; deCamp, Allan; Juraska, Michal; Fay, Michael P; Gilbert, Peter B
2017-04-01
Stepped wedge designs are increasingly commonplace and advantageous for cluster randomized trials when it is both unethical to assign placebo, and it is logistically difficult to allocate an intervention simultaneously to many clusters. We study marginal mean models fit with generalized estimating equations for assessing treatment effectiveness in stepped wedge cluster randomized trials. This approach has advantages over the more commonly used mixed models that (1) the population-average parameters have an important interpretation for public health applications and (2) they avoid untestable assumptions on latent variable distributions and avoid parametric assumptions about error distributions, therefore, providing more robust evidence on treatment effects. However, cluster randomized trials typically have a small number of clusters, rendering the standard generalized estimating equation sandwich variance estimator biased and highly variable and hence yielding incorrect inferences. We study the usual asymptotic generalized estimating equation inferences (i.e., using sandwich variance estimators and asymptotic normality) and four small-sample corrections to generalized estimating equation for stepped wedge cluster randomized trials and for parallel cluster randomized trials as a comparison. We show by simulation that the small-sample corrections provide improvement, with one correction appearing to provide at least nominal coverage even with only 10 clusters per group. These results demonstrate the viability of the marginal mean approach for both stepped wedge and parallel cluster randomized trials. We also study the comparative performance of the corrected methods for stepped wedge and parallel designs, and describe how the methods can accommodate interval censoring of individual failure times and incorporate semiparametric efficient estimators.
An Investigation of the Raudenbush (1988) Test for Studying Variance Heterogeneity.
ERIC Educational Resources Information Center
Harwell, Michael
1997-01-01
The meta-analytic method proposed by S. W. Raudenbush (1988) for studying variance heterogeneity was studied. Results of a Monte Carlo study indicate that the Type I error rate of the test is sensitive to even modestly platykurtic score distributions and to the ratio of study sample size to the number of studies. (SLD)
Newell, Felicity L.; Sheehan, James; Wood, Petra Bohall; Rodewald, Amanda D.; Buehler, David A.; Keyser, Patrick D.; Larkin, Jeffrey L.; Beachy, Tiffany A.; Bakermans, Marja H.; Boves, Than J.; Evans, Andrea; George, Gregory A.; McDermott, Molly E.; Perkins, Kelly A.; White, Matthew; Wigley, T. Bently
2013-01-01
Point counts are commonly used to assess changes in bird abundance, including analytical approaches such as distance sampling that estimate density. Point-count methods have come under increasing scrutiny because effects of detection probability and field error are difficult to quantify. For seven forest songbirds, we compared fixed-radii counts (50 m and 100 m) and density estimates obtained from distance sampling to known numbers of birds determined by territory mapping. We applied point-count analytic approaches to a typical forest management question and compared results to those obtained by territory mapping. We used a before–after control impact (BACI) analysis with a data set collected across seven study areas in the central Appalachians from 2006 to 2010. Using a 50-m fixed radius, variance in error was at least 1.5 times that of the other methods, whereas a 100-m fixed radius underestimated actual density by >3 territories per 10 ha for the most abundant species. Distance sampling improved accuracy and precision compared to fixed-radius counts, although estimates were affected by birds counted outside 10-ha units. In the BACI analysis, territory mapping detected an overall treatment effect for five of the seven species, and effects were generally consistent each year. In contrast, all point-count methods failed to detect two treatment effects due to variance and error in annual estimates. Overall, our results highlight the need for adequate sample sizes to reduce variance, and skilled observers to reduce the level of error in point-count data. Ultimately, the advantages and disadvantages of different survey methods should be considered in the context of overall study design and objectives, allowing for trade-offs among effort, accuracy, and power to detect treatment effects.
Analyzing thematic maps and mapping for accuracy
Rosenfield, G.H.
1982-01-01
Two problems which exist while attempting to test the accuracy of thematic maps and mapping are: (1) evaluating the accuracy of thematic content, and (2) evaluating the effects of the variables on thematic mapping. Statistical analysis techniques are applicable to both these problems and include techniques for sampling the data and determining their accuracy. In addition, techniques for hypothesis testing, or inferential statistics, are used when comparing the effects of variables. A comprehensive and valid accuracy test of a classification project, such as thematic mapping from remotely sensed data, includes the following components of statistical analysis: (1) sample design, including the sample distribution, sample size, size of the sample unit, and sampling procedure; and (2) accuracy estimation, including estimation of the variance and confidence limits. Careful consideration must be given to the minimum sample size necessary to validate the accuracy of a given. classification category. The results of an accuracy test are presented in a contingency table sometimes called a classification error matrix. Usually the rows represent the interpretation, and the columns represent the verification. The diagonal elements represent the correct classifications. The remaining elements of the rows represent errors by commission, and the remaining elements of the columns represent the errors of omission. For tests of hypothesis that compare variables, the general practice has been to use only the diagonal elements from several related classification error matrices. These data are arranged in the form of another contingency table. The columns of the table represent the different variables being compared, such as different scales of mapping. The rows represent the blocking characteristics, such as the various categories of classification. The values in the cells of the tables might be the counts of correct classification or the binomial proportions of these counts divided by either the row totals or the column totals from the original classification error matrices. In hypothesis testing, when the results of tests of multiple sample cases prove to be significant, some form of statistical test must be used to separate any results that differ significantly from the others. In the past, many analyses of the data in this error matrix were made by comparing the relative magnitudes of the percentage of correct classifications, for either individual categories, the entire map or both. More rigorous analyses have used data transformations and (or) two-way classification analysis of variance. A more sophisticated step of data analysis techniques would be to use the entire classification error matrices using the methods of discrete multivariate analysis or of multiviariate analysis of variance.
Meta-analysis with missing study-level sample variance data.
Chowdhry, Amit K; Dworkin, Robert H; McDermott, Michael P
2016-07-30
We consider a study-level meta-analysis with a normally distributed outcome variable and possibly unequal study-level variances, where the object of inference is the difference in means between a treatment and control group. A common complication in such an analysis is missing sample variances for some studies. A frequently used approach is to impute the weighted (by sample size) mean of the observed variances (mean imputation). Another approach is to include only those studies with variances reported (complete case analysis). Both mean imputation and complete case analysis are only valid under the missing-completely-at-random assumption, and even then the inverse variance weights produced are not necessarily optimal. We propose a multiple imputation method employing gamma meta-regression to impute the missing sample variances. Our method takes advantage of study-level covariates that may be used to provide information about the missing data. Through simulation studies, we show that multiple imputation, when the imputation model is correctly specified, is superior to competing methods in terms of confidence interval coverage probability and type I error probability when testing a specified group difference. Finally, we describe a similar approach to handling missing variances in cross-over studies. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.
Craft, Daniel F; Kry, Stephen F; Balter, Peter; Salehpour, Mohammad; Woodward, Wendy; Howell, Rebecca M
2018-04-01
Using 3D printing to fabricate patient-specific devices such as tissue compensators, boluses, and phantoms is inexpensive and relatively simple. However, most 3D printing materials have not been well characterized, including their radiologic tissue equivalence. The purposes of this study were to (a) determine the variance in Hounsfield Units (HU) for printed objects, (b) determine if HU varies over time, and (c) calculate the clinical dose uncertainty caused by these material variations. For a sample of 10 printed blocks each of PLA, NinjaFlex, ABS, and Cheetah, the average HU and physical density were tracked at initial printing and over the course of 5 weeks, a typical timeframe for a standard course of radiotherapy. After initial printing, half the blocks were stored in open boxes, the other half in sealed bags with desiccant. Variances in HU and density over time were evaluated for the four materials. Various clinical photon and electron beams were used to evaluate potential errors in clinical depth dose as a function of assumptions made during treatment planning. The clinical depth error was defined as the distance between the correctly calculated 90% isodose line and the 90% isodose line calculated using clinically reasonable, but simplified, assumptions. The average HU measurements of individual blocks of PLA, ABS, NinjaFlex, and Cheetah varied by as much as 121, 30, 178, and 30 HU, respectively. The HU variation over 5 weeks was much smaller for all materials. The magnitude of clinical depth errors depended strongly on the material, energy, and assumptions, but some were as large as 9.0 mm. If proper quality assurance steps are taken, 3D printed objects can be used accurately and effectively in radiation therapy. It is critically important, however, that the properties of any material being used in patient care be well understood and accounted for. © 2018 American Association of Physicists in Medicine.
Relationships of Measurement Error and Prediction Error in Observed-Score Regression
ERIC Educational Resources Information Center
Moses, Tim
2012-01-01
The focus of this paper is assessing the impact of measurement errors on the prediction error of an observed-score regression. Measures are presented and described for decomposing the linear regression's prediction error variance into parts attributable to the true score variance and the error variances of the dependent variable and the predictor…
Rast, Philippe; Hofer, Scott M.
2014-01-01
We investigated the power to detect variances and covariances in rates of change in the context of existing longitudinal studies using linear bivariate growth curve models. Power was estimated by means of Monte Carlo simulations. Our findings show that typical longitudinal study designs have substantial power to detect both variances and covariances among rates of change in a variety of cognitive, physical functioning, and mental health outcomes. We performed simulations to investigate the interplay among number and spacing of occasions, total duration of the study, effect size, and error variance on power and required sample size. The relation between growth rate reliability (GRR) and effect size to the sample size required to detect power ≥ .80 was non-linear, with rapidly decreasing sample sizes needed as GRR increases. The results presented here stand in contrast to previous simulation results and recommendations (Hertzog, Lindenberger, Ghisletta, & von Oertzen, 2006; Hertzog, von Oertzen, Ghisletta, & Lindenberger, 2008; von Oertzen, Ghisletta, & Lindenberger, 2010), which are limited due to confounds between study length and number of waves, error variance with GCR, and parameter values which are largely out of bounds of actual study values. Power to detect change is generally low in the early phases (i.e. first years) of longitudinal studies but can substantially increase if the design is optimized. We recommend additional assessments, including embedded intensive measurement designs, to improve power in the early phases of long-term longitudinal studies. PMID:24219544
NASA Technical Reports Server (NTRS)
Mozer, F. S.
1976-01-01
A computer program simulated the spectrum which resulted when a radar signal was transmitted into the ionosphere for a finite time and received for an equal finite interval. The spectrum derived from this signal is statistical in nature because the signal is scattered from the ionosphere, which is statistical in nature. Many estimates of any property of the ionosphere can be made. Their average value will approach the average property of the ionosphere which is being measured. Due to the statistical nature of the spectrum itself, the estimators will vary about this average. The square root of the variance about this average is called the standard deviation, an estimate of the error which exists in any particular radar measurement. In order to determine the feasibility of the space shuttle radar, the magnitude of these errors for measurements of physical interest must be understood.
Post-Modeling Histogram Matching of Maps Produced Using Regression Trees
Andrew J. Lister; Tonya W. Lister
2006-01-01
Spatial predictive models often use statistical techniques that in some way rely on averaging of values. Estimates from linear modeling are known to be susceptible to truncation of variance when the independent (predictor) variables are measured with error. A straightforward post-processing technique (histogram matching) for attempting to mitigate this effect is...
Meisner, Eric M; Hager, Gregory D; Ishman, Stacey L; Brown, David; Tunkel, David E; Ishii, Masaru
2013-11-01
To evaluate the accuracy of three-dimensional (3D) airway reconstructions obtained using quantitative endoscopy (QE). We developed this novel technique to reconstruct precise 3D representations of airway geometries from endoscopic video streams. This method, based on machine vision methodologies, uses a post-processing step of the standard videos obtained during routine laryngoscopy and bronchoscopy. We hypothesize that this method is precise and will generate assessment of airway size and shape similar to those obtained using computed tomography (CT). This study was approved by the institutional review board (IRB). We analyzed video sequences from pediatric patients receiving rigid bronchoscopy. We generated 3D scaled airway models of the subglottis, trachea, and carina using QE. These models were compared to 3D airway models generated from CT. We used the CT data as the gold standard measure of airway size, and used a mixed linear model to estimate the average error in cross-sectional area and effective diameter for QE. The average error in cross sectional area (area sliced perpendicular to the long axis of the airway) was 7.7 mm(2) (variance 33.447 mm(4)). The average error in effective diameter was 0.38775 mm (variance 2.45 mm(2)), approximately 9% error. Our pilot study suggests that QE can be used to generate precise 3D reconstructions of airways. This technique is atraumatic, does not require ionizing radiation, and integrates easily into standard airway assessment protocols. We conjecture that this technology will be useful for staging airway disease and assessing surgical outcomes. Copyright © 2013 The American Laryngological, Rhinological and Otological Society, Inc.
Serafini, Kelly; Malin-Mayor, Bo; Nich, Charla; Hunkele, Karen; Carroll, Kathleen M
2016-03-01
The Positive and Negative Affect Schedule (PANAS) is a widely used measure of affect. A comprehensive psychometric evaluation among substance users, however, has not been published. To examine the psychometric properties of the PANAS in a sample of outpatient treatment substance users. We used pooled data from four randomized clinical trials (N = 416; 34% female, 48% African American). A confirmatory factor analysis indicated adequate support for a two-factor correlated model comprised of Positive Affect and Negative Affect with correlated item errors (Comparative Fit Index = 0.93, Root Mean Square Error of Approximation = 0.07, χ(2) = 478.93, df = 156). Cronbach's α indicated excellent internal consistency for both factors (0.90 and 0.91, respectively). The PANAS factors had good convergence and discriminability (Composite Reliability > 0.7; Maximum Shared Variance < Average Variance Extracted). A comparison from baseline to Week 1 indicated acceptable test-retest reliability (Positive Affect = 0.80, Negative Affect = 0.76). Concurrent and discriminant validity were demonstrated with correlations with the Brief Symptom Inventory and Addiction Severity Index. The PANAS scores were also significantly correlated with treatment outcomes (e.g. Positive Affect was associated with the maximum days of consecutive abstinence from primary substance of abuse, r = 0.16, p = 0.001). Our data suggest that the psychometric properties of the PANAS are retained in substance using populations. Although several studies have focused on the role of Negative Affect, our findings suggest that Positive Affect may also be an important factor in substance use treatment outcomes.
Serafini, Kelly; Malin-Mayor, Bo; Nich, Charla; Hunkele, Karen; Carroll, Kathleen M.
2016-01-01
Background The Positive and Negative Affect Schedule (PANAS) is a widely used measure of affect, and a comprehensive psychometric evaluation has never been conducted among substance users. Objective To examine the psychometric properties of the PANAS in a sample of outpatient treatment substance users. Methods We used pooled data from four randomized clinical trials (N = 416; 34% female, 48% African American). Results A confirmatory factor analysis indicated adequate support for a two-factor correlated model comprised of Positive Affect and Negative Affect with correlated item errors (Comparative Fit Index = .93, Root Mean Square Error of Approximation = .07, χ2 = 478.93, df = 156). Cronbach’s α indicated excellent internal consistency for both factors (.90 and .91, respectively). The PANAS factors had good convergence and discriminability (Composite Reliability >.7; Maximum Shared Variance < Average Variance Extracted). A comparison from baseline to Week 1 indicated acceptable test-retest reliability (Positive Affect = .80, Negative Affect = .76). Concurrent and discriminant validity were demonstrated with correlations with the Brief Symptom Inventory and Addiction Severity Index. The PANAS scores were also significantly correlated with treatment outcomes (e.g., Positive Affect was associated with the maximum days of consecutive abstinence from primary substance of abuse, r = .16, p = .001). Conclusion Our data suggest that the psychometric properties of the PANAS are retained in substance using populations. Although several studies have focused on the role of Negative Affect, our findings suggest that Positive Affect may also be an important factor in substance use treatment outcomes. PMID:26905228
NASA Astrophysics Data System (ADS)
El-Diasty, M.; El-Rabbany, A.; Pagiatakis, S.
2007-11-01
We examine the effect of varying the temperature points on MEMS inertial sensors' noise models using Allan variance and least-squares spectral analysis (LSSA). Allan variance is a method of representing root-mean-square random drift error as a function of averaging times. LSSA is an alternative to the classical Fourier methods and has been applied successfully by a number of researchers in the study of the noise characteristics of experimental series. Static data sets are collected at different temperature points using two MEMS-based IMUs, namely MotionPakII and Crossbow AHRS300CC. The performance of the two MEMS inertial sensors is predicted from the Allan variance estimation results at different temperature points and the LSSA is used to study the noise characteristics and define the sensors' stochastic model parameters. It is shown that the stochastic characteristics of MEMS-based inertial sensors can be identified using Allan variance estimation and LSSA and the sensors' stochastic model parameters are temperature dependent. Also, the Kaiser window FIR low-pass filter is used to investigate the effect of de-noising stage on the stochastic model. It is shown that the stochastic model is also dependent on the chosen cut-off frequency.
Sangnawakij, Patarawan; Böhning, Dankmar; Adams, Stephen; Stanton, Michael; Holling, Heinz
2017-04-30
Statistical inference for analyzing the results from several independent studies on the same quantity of interest has been investigated frequently in recent decades. Typically, any meta-analytic inference requires that the quantity of interest is available from each study together with an estimate of its variability. The current work is motivated by a meta-analysis on comparing two treatments (thoracoscopic and open) of congenital lung malformations in young children. Quantities of interest include continuous end-points such as length of operation or number of chest tube days. As studies only report mean values (and no standard errors or confidence intervals), the question arises how meta-analytic inference can be developed. We suggest two methods to estimate study-specific variances in such a meta-analysis, where only sample means and sample sizes are available in the treatment arms. A general likelihood ratio test is derived for testing equality of variances in two groups. By means of simulation studies, the bias and estimated standard error of the overall mean difference from both methodologies are evaluated and compared with two existing approaches: complete study analysis only and partial variance information. The performance of the test is evaluated in terms of type I error. Additionally, we illustrate these methods in the meta-analysis on comparing thoracoscopic and open surgery for congenital lung malformations and in a meta-analysis on the change in renal function after kidney donation. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.
A Note on Sample Size and Solution Propriety for Confirmatory Factor Analytic Models
ERIC Educational Resources Information Center
Jackson, Dennis L.; Voth, Jennifer; Frey, Marc P.
2013-01-01
Determining an appropriate sample size for use in latent variable modeling techniques has presented ongoing challenges to researchers. In particular, small sample sizes are known to present concerns over sampling error for the variances and covariances on which model estimation is based, as well as for fit indexes and convergence failures. The…
Crop area estimation based on remotely-sensed data with an accurate but costly subsample
NASA Technical Reports Server (NTRS)
Gunst, R. F.
1983-01-01
Alternatives to sampling-theory stratified and regression estimators of crop production and timber biomass were examined. An alternative estimator which is viewed as especially promising is the errors-in-variable regression estimator. Investigations established the need for caution with this estimator when the ratio of two error variances is not precisely known.
Analytic score distributions for a spatially continuous tridirectional Monte Carol transport problem
DOE Office of Scientific and Technical Information (OSTI.GOV)
Booth, T.E.
1996-01-01
The interpretation of the statistical error estimates produced by Monte Carlo transport codes is still somewhat of an art. Empirically, there are variance reduction techniques whose error estimates are almost always reliable, and there are variance reduction techniques whose error estimates are often unreliable. Unreliable error estimates usually result from inadequate large-score sampling from the score distribution`s tail. Statisticians believe that more accurate confidence interval statements are possible if the general nature of the score distribution can be characterized. Here, the analytic score distribution for the exponential transform applied to a simple, spatially continuous Monte Carlo transport problem is provided.more » Anisotropic scattering and implicit capture are included in the theory. In large part, the analytic score distributions that are derived provide the basis for the ten new statistical quality checks in MCNP.« less
Replica approach to mean-variance portfolio optimization
NASA Astrophysics Data System (ADS)
Varga-Haszonits, Istvan; Caccioli, Fabio; Kondor, Imre
2016-12-01
We consider the problem of mean-variance portfolio optimization for a generic covariance matrix subject to the budget constraint and the constraint for the expected return, with the application of the replica method borrowed from the statistical physics of disordered systems. We find that the replica symmetry of the solution does not need to be assumed, but emerges as the unique solution of the optimization problem. We also check the stability of this solution and find that the eigenvalues of the Hessian are positive for r = N/T < 1, where N is the dimension of the portfolio and T the length of the time series used to estimate the covariance matrix. At the critical point r = 1 a phase transition is taking place. The out of sample estimation error blows up at this point as 1/(1 - r), independently of the covariance matrix or the expected return, displaying the universality not only of the critical exponent, but also the critical point. As a conspicuous illustration of the dangers of in-sample estimates, the optimal in-sample variance is found to vanish at the critical point inversely proportional to the divergent estimation error.
Knopman, Debra S.; Voss, Clifford I.
1987-01-01
The spatial and temporal variability of sensitivities has a significant impact on parameter estimation and sampling design for studies of solute transport in porous media. Physical insight into the behavior of sensitivities is offered through an analysis of analytically derived sensitivities for the one-dimensional form of the advection-dispersion equation. When parameters are estimated in regression models of one-dimensional transport, the spatial and temporal variability in sensitivities influences variance and covariance of parameter estimates. Several principles account for the observed influence of sensitivities on parameter uncertainty. (1) Information about a physical parameter may be most accurately gained at points in space and time with a high sensitivity to the parameter. (2) As the distance of observation points from the upstream boundary increases, maximum sensitivity to velocity during passage of the solute front increases and the consequent estimate of velocity tends to have lower variance. (3) The frequency of sampling must be “in phase” with the S shape of the dispersion sensitivity curve to yield the most information on dispersion. (4) The sensitivity to the dispersion coefficient is usually at least an order of magnitude less than the sensitivity to velocity. (5) The assumed probability distribution of random error in observations of solute concentration determines the form of the sensitivities. (6) If variance in random error in observations is large, trends in sensitivities of observation points may be obscured by noise and thus have limited value in predicting variance in parameter estimates among designs. (7) Designs that minimize the variance of one parameter may not necessarily minimize the variance of other parameters. (8) The time and space interval over which an observation point is sensitive to a given parameter depends on the actual values of the parameters in the underlying physical system.
Topping, David J.; Rubin, David M.; Wright, Scott A.; Melis, Theodore S.
2011-01-01
Several common methods for measuring suspended-sediment concentration in rivers in the United States use depth-integrating samplers to collect a velocity-weighted suspended-sediment sample in a subsample of a river cross section. Because depth-integrating samplers are always moving through the water column as they collect a sample, and can collect only a limited volume of water and suspended sediment, they collect only minimally time-averaged data. Four sources of error exist in the field use of these samplers: (1) bed contamination, (2) pressure-driven inrush, (3) inadequate sampling of the cross-stream spatial structure in suspended-sediment concentration, and (4) inadequate time averaging. The first two of these errors arise from misuse of suspended-sediment samplers, and the third has been the subject of previous study using data collected in the sand-bedded Middle Loup River in Nebraska. Of these four sources of error, the least understood source of error arises from the fact that depth-integrating samplers collect only minimally time-averaged data. To evaluate this fourth source of error, we collected suspended-sediment data between 1995 and 2007 at four sites on the Colorado River in Utah and Arizona, using a P-61 suspended-sediment sampler deployed in both point- and one-way depth-integrating modes, and D-96-A1 and D-77 bag-type depth-integrating suspended-sediment samplers. These data indicate that the minimal duration of time averaging during standard field operation of depth-integrating samplers leads to an error that is comparable in magnitude to that arising from inadequate sampling of the cross-stream spatial structure in suspended-sediment concentration. This random error arising from inadequate time averaging is positively correlated with grain size and does not largely depend on flow conditions or, for a given size class of suspended sediment, on elevation above the bed. Averaging over time scales >1 minute is the likely minimum duration required to result in substantial decreases in this error. During standard two-way depth integration, a depth-integrating suspended-sediment sampler collects a sample of the water-sediment mixture during two transits at each vertical in a cross section: one transit while moving from the water surface to the bed, and another transit while moving from the bed to the water surface. As the number of transits is doubled at an individual vertical, this error is reduced by ~30 percent in each size class of suspended sediment. For a given size class of suspended sediment, the error arising from inadequate sampling of the cross-stream spatial structure in suspended-sediment concentration depends only on the number of verticals collected, whereas the error arising from inadequate time averaging depends on both the number of verticals collected and the number of transits collected at each vertical. Summing these two errors in quadrature yields a total uncertainty in an equal-discharge-increment (EDI) or equal-width-increment (EWI) measurement of the time-averaged velocity-weighted suspended-sediment concentration in a river cross section (exclusive of any laboratory-processing errors). By virtue of how the number of verticals and transits influences the two individual errors within this total uncertainty, the error arising from inadequate time averaging slightly dominates that arising from inadequate sampling of the cross-stream spatial structure in suspended-sediment concentration. Adding verticals to an EDI or EWI measurement is slightly more effective in reducing the total uncertainty than adding transits only at each vertical, because a new vertical contributes both temporal and spatial information. However, because collection of depth-integrated samples at more transits at each vertical is generally easier and faster than at more verticals, addition of a combination of verticals and transits is likely a more practical approach to reducing the total uncertainty in most field situatio
Performance of correlation receivers in the presence of impulse noise.
NASA Technical Reports Server (NTRS)
Moore, J. D.; Houts, R. C.
1972-01-01
An impulse noise model, which assumes that each noise burst contains a randomly weighted version of a basic waveform, is used to derive the performance equations for a correlation receiver. The expected number of bit errors per noise burst is expressed as a function of the average signal energy, signal-set correlation coefficient, bit time, noise-weighting-factor variance and probability density function, and a time range function which depends on the crosscorrelation of the signal-set basis functions and the noise waveform. Unlike the performance results for additive white Gaussian noise, it is shown that the error performance for impulse noise is affected by the choice of signal-set basis function, and that Orthogonal signaling is not equivalent to On-Off signaling with the same average energy. Furthermore, it is demonstrated that the correlation-receiver error performance can be improved by inserting a properly specified nonlinear device prior to the receiver input.
Automated Grouping of Opportunity Rover Alpha Particle X-Ray Spectrometer Compositional Data
NASA Technical Reports Server (NTRS)
VanBommel, S. J.; Gellert, R.; Clark, B. C.; Ming, D. W.; Mittlefehldt, D. W.; Schroder, C.; Yen, A. S.
2016-01-01
The Alpha Particle X-ray Spectrometer (APXS) conducts high-precision in situ measurements of rocks and soils on both active NASA Mars rovers. Since 2004 the rover Opportunity has acquired around 440 unique APXS measurements, including a wide variety of compositions, during its 42+ kilometers traverse across several geological formations. Here we discuss an analytical comparison algorithm providing a means to cluster samples due to compositional similarity and the resulting automated classification scheme. Due to the inherent variance of elements in the APXS data set, each element has an associated weight that is inversely proportional to the variance. Thus, the more consistent the abundance of an element in the data set, the more it contributes to the classification. All 16 elements standard to the APXS data set are considered. Careful attention is also given to the errors associated with the composition measured by the APXS - larger uncertainties reduce the weighting of the element accordingly. The comparison of two targets, i and j, generates a similarity score, S(sub ij). This score is immediately comparable to an average ratio across all elements if one assumes standard weighted uncertainty. The algorithm facilitates the classification of APXS targets by chemistry alone - independent of target appearance and geological context which can be added later as a consistency check. For the N targets considered, a N by N hollow matrix, S, is generated where S = S(sup T). The average relation score, S(sub av), for target N(sub i) is simply the average of column i of S. A large S(sub av) is indicative of a unique sample. In such an instance any targets with a low comparison score can be classified alike. The threshold between classes requires careful consideration. Applying the algorithm to recent Marathon Valley targets indicates similarities with Burns formation and average-Mars-like rocks encountered earlier at Endeavour Crater as well as a new class of felsic rocks.
On the error in crop acreage estimation using satellite (LANDSAT) data
NASA Technical Reports Server (NTRS)
Chhikara, R. (Principal Investigator)
1983-01-01
The problem of crop acreage estimation using satellite data is discussed. Bias and variance of a crop proportion estimate in an area segment obtained from the classification of its multispectral sensor data are derived as functions of the means, variances, and covariance of error rates. The linear discriminant analysis and the class proportion estimation for the two class case are extended to include a third class of measurement units, where these units are mixed on ground. Special attention is given to the investigation of mislabeling in training samples and its effect on crop proportion estimation. It is shown that the bias and variance of the estimate of a specific crop acreage proportion increase as the disparity in mislabeling rates between two classes increases. Some interaction is shown to take place, causing the bias and the variance to decrease at first and then to increase, as the mixed unit class varies in size from 0 to 50 percent of the total area segment.
NASA Technical Reports Server (NTRS)
Bell, Thomas L.; Abdullah, A.; Martin, Russell L.; North, Gerald R.
1990-01-01
Estimates of monthly average rainfall based on satellite observations from a low earth orbit will differ from the true monthly average because the satellite observes a given area only intermittently. This sampling error inherent in satellite monitoring of rainfall would occur even if the satellite instruments could measure rainfall perfectly. The size of this error is estimated for a satellite system being studied at NASA, the Tropical Rainfall Measuring Mission (TRMM). First, the statistical description of rainfall on scales from 1 to 1000 km is examined in detail, based on rainfall data from the Global Atmospheric Research Project Atlantic Tropical Experiment (GATE). A TRMM-like satellite is flown over a two-dimensional time-evolving simulation of rainfall using a stochastic model with statistics tuned to agree with GATE statistics. The distribution of sampling errors found from many months of simulated observations is found to be nearly normal, even though the distribution of area-averaged rainfall is far from normal. For a range of orbits likely to be employed in TRMM, sampling error is found to be less than 10 percent of the mean for rainfall averaged over a 500 x 500 sq km area.
ERIC Educational Resources Information Center
Longford, Nicholas T.
Large scale surveys usually employ a complex sampling design and as a consequence, no standard methods for estimation of the standard errors associated with the estimates of population means are available. Resampling methods, such as jackknife or bootstrap, are often used, with reference to their properties of robustness and reduction of bias. A…
Gelbrich, Bianca; Frerking, Carolin; Weiss, Sandra; Schwerdt, Sebastian; Stellzig-Eisenhauer, Angelika; Tausche, Eve; Gelbrich, Götz
2015-01-01
Forensic age estimation in living adolescents is based on several methods, e.g. the assessment of skeletal and dental maturation. Combination of several methods is mandatory, since age estimates from a single method are too imprecise due to biological variability. The correlation of the errors of the methods being combined must be known to calculate the precision of combined age estimates. To examine the correlation of the errors of the hand and the third molar method and to demonstrate how to calculate the combined age estimate. Clinical routine radiographs of the hand and dental panoramic images of 383 patients (aged 7.8-19.1 years, 56% female) were assessed. Lack of correlation (r = -0.024, 95% CI = -0.124 to + 0.076, p = 0.64) allows calculating the combined age estimate as the weighted average of the estimates from hand bones and third molars. Combination improved the standard deviations of errors (hand = 0.97, teeth = 1.35 years) to 0.79 years. Uncorrelated errors of the age estimates obtained from both methods allow straightforward determination of the common estimate and its variance. This is also possible when reference data for the hand and the third molar method are established independently from each other, using different samples.
Survival analysis with error-prone time-varying covariates: a risk set calibration approach
Liao, Xiaomei; Zucker, David M.; Li, Yi; Spiegelman, Donna
2010-01-01
Summary Occupational, environmental, and nutritional epidemiologists are often interested in estimating the prospective effect of time-varying exposure variables such as cumulative exposure or cumulative updated average exposure, in relation to chronic disease endpoints such as cancer incidence and mortality. From exposure validation studies, it is apparent that many of the variables of interest are measured with moderate to substantial error. Although the ordinary regression calibration approach is approximately valid and efficient for measurement error correction of relative risk estimates from the Cox model with time-independent point exposures when the disease is rare, it is not adaptable for use with time-varying exposures. By re-calibrating the measurement error model within each risk set, a risk set regression calibration method is proposed for this setting. An algorithm for a bias-corrected point estimate of the relative risk using an RRC approach is presented, followed by the derivation of an estimate of its variance, resulting in a sandwich estimator. Emphasis is on methods applicable to the main study/external validation study design, which arises in important applications. Simulation studies under several assumptions about the error model were carried out, which demonstrated the validity and efficiency of the method in finite samples. The method was applied to a study of diet and cancer from Harvard’s Health Professionals Follow-up Study (HPFS). PMID:20486928
A stream-gaging network analysis for the 7-day, 10-year annual low flow in New Hampshire streams
Flynn, Robert H.
2003-01-01
The 7-day, 10-year (7Q10) low-flow-frequency statistic is a widely used measure of surface-water availability in New Hampshire. Regression equations and basin-characteristic digital data sets were developed to help water-resource managers determine surface-water resources during periods of low flow in New Hampshire streams. These regression equations and data sets were developed to estimate streamflow statistics for the annual and seasonal low-flow-frequency, and period-of-record and seasonal period-of-record flow durations. generalized-least-squares (GLS) regression methods were used to develop the annual 7Q10 low-flow-frequency regression equation from 60 continuous-record stream-gaging stations in New Hampshire and in neighboring States. In the regression equation, the dependent variables were the annual 7Q10 flows at the 60 stream-gaging stations. The independent (or predictor) variables were objectively selected characteristics of the drainage basins that contribute flow to those stations. In contrast to ordinary-least-squares (OLS) regression analysis, GLS-developed estimating equations account for differences in length of record and spatial correlations among the flow-frequency statistics at the various stations.A total of 93 measurable drainage-basin characteristics were candidate independent variables. On the basis of several statistical parameters that were used to evaluate which combination of basin characteristics contribute the most to the predictive power of the equations, three drainage-basin characteristics were determined to be statistically significant predictors of the annual 7Q10: (1) total drainage area, (2) mean summer stream-gaging station precipitation from 1961 to 90, and (3) average mean annual basinwide temperature from 1961 to 1990.To evaluate the effectiveness of the stream-gaging network in providing regional streamflow data for the annual 7Q10, the computer program GLSNET (generalized-least-squares NETwork) was used to analyze the network by application of GLS regression between streamflow and the climatic and basin characteristics of the drainage basin upstream from each stream-gaging station. Improvement to the predictive ability of the regression equations developed for the network analyses is measured by the reduction in the average sampling-error variance, and can be achieved by collecting additional streamflow data at existing stations. The predictive ability of the regression equations is enhanced even further with the addition of new stations to the network. Continued data collection at unregulated stream-gaging stations with less than 14 years of record resulted in the greatest cost-weighted reduction to the average sampling-error variance of the annual 7Q10 regional regression equation. The addition of new stations in basins with underrepresented values for the independent variables of the total drainage area, average mean annual basinwide temperature, or mean summer stream-gaging station precipitation in the annual 7Q10 regression equation yielded a much greater cost-weighted reduction to the average sampling-error variance than when more data were collected at existing unregulated stations. To maximize the regional information obtained from the stream-gaging network for the annual 7Q10, ranking of the streamflow data can be used to determine whether an active station should be continued or if a new or discontinued station should be activated for streamflow data collection. Thus, this network analysis can help determine the costs and benefits of continuing the operation of a particular station or activating a new station at another location to predict the 7Q10 at ungaged stream reaches. The decision to discontinue an existing station or activate a new station, however, must also consider its contribution to other water-resource analyses such as flood management, water quality, or trends in land use or climatic change.
An application of the LC-LSTM framework to the self-esteem instability case.
Alessandri, Guido; Vecchione, Michele; Donnellan, Brent M; Tisak, John
2013-10-01
The present research evaluates the stability of self-esteem as assessed by a daily version of the Rosenberg (Society and the adolescent self-image, Princeton University Press, Princeton, 1965) general self-esteem scale (RGSE). The scale was administered to 391 undergraduates for five consecutive days. The longitudinal data were analyzed using the integrated LC-LSTM framework that allowed us to evaluate: (1) the measurement invariance of the RGSE, (2) its stability and change across the 5-day assessment period, (3) the amount of variance attributable to stable and transitory latent factors, and (4) the criterion-related validity of these factors. Results provided evidence for measurement invariance, mean-level stability, and rank-order stability of daily self-esteem. Latent state-trait analyses revealed that variances in scores of the RGSE can be decomposed into six components: stable self-esteem (40 %), ephemeral (or temporal-state) variance (36 %), stable negative method variance (9 %), stable positive method variance (4 %), specific variance (1 %) and random error variance (10 %). Moreover, latent factors associated with daily self-esteem were associated with measures of depression, implicit self-esteem, and grade point average.
Blinded sample size re-estimation in three-arm trials with 'gold standard' design.
Mütze, Tobias; Friede, Tim
2017-10-15
In this article, we study blinded sample size re-estimation in the 'gold standard' design with internal pilot study for normally distributed outcomes. The 'gold standard' design is a three-arm clinical trial design that includes an active and a placebo control in addition to an experimental treatment. We focus on the absolute margin approach to hypothesis testing in three-arm trials at which the non-inferiority of the experimental treatment and the assay sensitivity are assessed by pairwise comparisons. We compare several blinded sample size re-estimation procedures in a simulation study assessing operating characteristics including power and type I error. We find that sample size re-estimation based on the popular one-sample variance estimator results in overpowered trials. Moreover, sample size re-estimation based on unbiased variance estimators such as the Xing-Ganju variance estimator results in underpowered trials, as it is expected because an overestimation of the variance and thus the sample size is in general required for the re-estimation procedure to eventually meet the target power. To overcome this problem, we propose an inflation factor for the sample size re-estimation with the Xing-Ganju variance estimator and show that this approach results in adequately powered trials. Because of favorable features of the Xing-Ganju variance estimator such as unbiasedness and a distribution independent of the group means, the inflation factor does not depend on the nuisance parameter and, therefore, can be calculated prior to a trial. Moreover, we prove that the sample size re-estimation based on the Xing-Ganju variance estimator does not bias the effect estimate. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.
Technical Note: Introduction of variance component analysis to setup error analysis in radiotherapy
DOE Office of Scientific and Technical Information (OSTI.GOV)
Matsuo, Yukinori, E-mail: ymatsuo@kuhp.kyoto-u.ac.
Purpose: The purpose of this technical note is to introduce variance component analysis to the estimation of systematic and random components in setup error of radiotherapy. Methods: Balanced data according to the one-factor random effect model were assumed. Results: Analysis-of-variance (ANOVA)-based computation was applied to estimate the values and their confidence intervals (CIs) for systematic and random errors and the population mean of setup errors. The conventional method overestimates systematic error, especially in hypofractionated settings. The CI for systematic error becomes much wider than that for random error. The ANOVA-based estimation can be extended to a multifactor model considering multiplemore » causes of setup errors (e.g., interpatient, interfraction, and intrafraction). Conclusions: Variance component analysis may lead to novel applications to setup error analysis in radiotherapy.« less
Statistical methods for biodosimetry in the presence of both Berkson and classical measurement error
NASA Astrophysics Data System (ADS)
Miller, Austin
In radiation epidemiology, the true dose received by those exposed cannot be assessed directly. Physical dosimetry uses a deterministic function of the source term, distance and shielding to estimate dose. For the atomic bomb survivors, the physical dosimetry system is well established. The classical measurement errors plaguing the location and shielding inputs to the physical dosimetry system are well known. Adjusting for the associated biases requires an estimate for the classical measurement error variance, for which no data-driven estimate exists. In this case, an instrumental variable solution is the most viable option to overcome the classical measurement error indeterminacy. Biological indicators of dose may serve as instrumental variables. Specification of the biodosimeter dose-response model requires identification of the radiosensitivity variables, for which we develop statistical definitions and variables. More recently, researchers have recognized Berkson error in the dose estimates, introduced by averaging assumptions for many components in the physical dosimetry system. We show that Berkson error induces a bias in the instrumental variable estimate of the dose-response coefficient, and then address the estimation problem. This model is specified by developing an instrumental variable mixed measurement error likelihood function, which is then maximized using a Monte Carlo EM Algorithm. These methods produce dose estimates that incorporate information from both physical and biological indicators of dose, as well as the first instrumental variable based data-driven estimate for the classical measurement error variance.
Liu, Xiaofeng Steven
2011-05-01
The use of covariates is commonly believed to reduce the unexplained error variance and the standard error for the comparison of treatment means, but the reduction in the standard error is neither guaranteed nor uniform over different sample sizes. The covariate mean differences between the treatment conditions can inflate the standard error of the covariate-adjusted mean difference and can actually produce a larger standard error for the adjusted mean difference than that for the unadjusted mean difference. When the covariate observations are conceived of as randomly varying from one study to another, the covariate mean differences can be related to a Hotelling's T(2) . Using this Hotelling's T(2) statistic, one can always find a minimum sample size to achieve a high probability of reducing the standard error and confidence interval width for the adjusted mean difference. ©2010 The British Psychological Society.
Bernard R. Parresol
1993-01-01
In the context of forest modeling, it is often reasonable to assume a multiplicative heteroscedastic error structure to the data. Under such circumstances ordinary least squares no longer provides minimum variance estimates of the model parameters. Through study of the error structure, a suitable error variance model can be specified and its parameters estimated. This...
Linear and Order Statistics Combiners for Pattern Classification
NASA Technical Reports Server (NTRS)
Tumer, Kagan; Ghosh, Joydeep; Lau, Sonie (Technical Monitor)
2001-01-01
Several researchers have experimentally shown that substantial improvements can be obtained in difficult pattern recognition problems by combining or integrating the outputs of multiple classifiers. This chapter provides an analytical framework to quantify the improvements in classification results due to combining. The results apply to both linear combiners and order statistics combiners. We first show that to a first order approximation, the error rate obtained over and above the Bayes error rate, is directly proportional to the variance of the actual decision boundaries around the Bayes optimum boundary. Combining classifiers in output space reduces this variance, and hence reduces the 'added' error. If N unbiased classifiers are combined by simple averaging. the added error rate can be reduced by a factor of N if the individual errors in approximating the decision boundaries are uncorrelated. Expressions are then derived for linear combiners which are biased or correlated, and the effect of output correlations on ensemble performance is quantified. For order statistics based non-linear combiners, we derive expressions that indicate how much the median, the maximum and in general the i-th order statistic can improve classifier performance. The analysis presented here facilitates the understanding of the relationships among error rates, classifier boundary distributions, and combining in output space. Experimental results on several public domain data sets are provided to illustrate the benefits of combining and to support the analytical results.
Impact of multicollinearity on small sample hydrologic regression models
NASA Astrophysics Data System (ADS)
Kroll, Charles N.; Song, Peter
2013-06-01
Often hydrologic regression models are developed with ordinary least squares (OLS) procedures. The use of OLS with highly correlated explanatory variables produces multicollinearity, which creates highly sensitive parameter estimators with inflated variances and improper model selection. It is not clear how to best address multicollinearity in hydrologic regression models. Here a Monte Carlo simulation is developed to compare four techniques to address multicollinearity: OLS, OLS with variance inflation factor screening (VIF), principal component regression (PCR), and partial least squares regression (PLS). The performance of these four techniques was observed for varying sample sizes, correlation coefficients between the explanatory variables, and model error variances consistent with hydrologic regional regression models. The negative effects of multicollinearity are magnified at smaller sample sizes, higher correlations between the variables, and larger model error variances (smaller R2). The Monte Carlo simulation indicates that if the true model is known, multicollinearity is present, and the estimation and statistical testing of regression parameters are of interest, then PCR or PLS should be employed. If the model is unknown, or if the interest is solely on model predictions, is it recommended that OLS be employed since using more complicated techniques did not produce any improvement in model performance. A leave-one-out cross-validation case study was also performed using low-streamflow data sets from the eastern United States. Results indicate that OLS with stepwise selection generally produces models across study regions with varying levels of multicollinearity that are as good as biased regression techniques such as PCR and PLS.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Vidal-Codina, F., E-mail: fvidal@mit.edu; Nguyen, N.C., E-mail: cuongng@mit.edu; Giles, M.B., E-mail: mike.giles@maths.ox.ac.uk
We present a model and variance reduction method for the fast and reliable computation of statistical outputs of stochastic elliptic partial differential equations. Our method consists of three main ingredients: (1) the hybridizable discontinuous Galerkin (HDG) discretization of elliptic partial differential equations (PDEs), which allows us to obtain high-order accurate solutions of the governing PDE; (2) the reduced basis method for a new HDG discretization of the underlying PDE to enable real-time solution of the parameterized PDE in the presence of stochastic parameters; and (3) a multilevel variance reduction method that exploits the statistical correlation among the different reduced basismore » approximations and the high-fidelity HDG discretization to accelerate the convergence of the Monte Carlo simulations. The multilevel variance reduction method provides efficient computation of the statistical outputs by shifting most of the computational burden from the high-fidelity HDG approximation to the reduced basis approximations. Furthermore, we develop a posteriori error estimates for our approximations of the statistical outputs. Based on these error estimates, we propose an algorithm for optimally choosing both the dimensions of the reduced basis approximations and the sizes of Monte Carlo samples to achieve a given error tolerance. We provide numerical examples to demonstrate the performance of the proposed method.« less
Vallejo, Guillermo; Ato, Manuel; Fernández García, Paula; Livacic Rojas, Pablo E; Tuero Herrero, Ellián
2016-08-01
S. Usami (2014) describes a method to realistically determine sample size in longitudinal research using a multilevel model. The present research extends the aforementioned work to situations where it is likely that the assumption of homogeneity of the errors across groups is not met and the error term does not follow a scaled identity covariance structure. For this purpose, we followed a procedure based on transforming the variance components of the linear growth model and the parameter related to the treatment effect into specific and easily understandable indices. At the same time, we provide the appropriate statistical machinery for researchers to use when data loss is unavoidable, and changes in the expected value of the observed responses are not linear. The empirical powers based on unknown variance components were virtually the same as the theoretical powers derived from the use of statistically processed indexes. The main conclusion of the study is the accuracy of the proposed method to calculate sample size in the described situations with the stipulated power criteria.
Lu, Xinjiang; Liu, Wenbo; Zhou, Chuang; Huang, Minghui
2017-06-13
The least-squares support vector machine (LS-SVM) is a popular data-driven modeling method and has been successfully applied to a wide range of applications. However, it has some disadvantages, including being ineffective at handling non-Gaussian noise as well as being sensitive to outliers. In this paper, a robust LS-SVM method is proposed and is shown to have more reliable performance when modeling a nonlinear system under conditions where Gaussian or non-Gaussian noise is present. The construction of a new objective function allows for a reduction of the mean of the modeling error as well as the minimization of its variance, and it does not constrain the mean of the modeling error to zero. This differs from the traditional LS-SVM, which uses a worst-case scenario approach in order to minimize the modeling error and constrains the mean of the modeling error to zero. In doing so, the proposed method takes the modeling error distribution information into consideration and is thus less conservative and more robust in regards to random noise. A solving method is then developed in order to determine the optimal parameters for the proposed robust LS-SVM. An additional analysis indicates that the proposed LS-SVM gives a smaller weight to a large-error training sample and a larger weight to a small-error training sample, and is thus more robust than the traditional LS-SVM. The effectiveness of the proposed robust LS-SVM is demonstrated using both artificial and real life cases.
Estimation variance bounds of importance sampling simulations in digital communication systems
NASA Technical Reports Server (NTRS)
Lu, D.; Yao, K.
1991-01-01
In practical applications of importance sampling (IS) simulation, two basic problems are encountered, that of determining the estimation variance and that of evaluating the proper IS parameters needed in the simulations. The authors derive new upper and lower bounds on the estimation variance which are applicable to IS techniques. The upper bound is simple to evaluate and may be minimized by the proper selection of the IS parameter. Thus, lower and upper bounds on the improvement ratio of various IS techniques relative to the direct Monte Carlo simulation are also available. These bounds are shown to be useful and computationally simple to obtain. Based on the proposed technique, one can readily find practical suboptimum IS parameters. Numerical results indicate that these bounding techniques are useful for IS simulations of linear and nonlinear communication systems with intersymbol interference in which bit error rate and IS estimation variances cannot be obtained readily using prior techniques.
[Practical aspects regarding sample size in clinical research].
Vega Ramos, B; Peraza Yanes, O; Herrera Correa, G; Saldívar Toraya, S
1996-01-01
The knowledge of the right sample size let us to be sure if the published results in medical papers had a suitable design and a proper conclusion according to the statistics analysis. To estimate the sample size we must consider the type I error, type II error, variance, the size of the effect, significance and power of the test. To decide what kind of mathematics formula will be used, we must define what kind of study we have, it means if its a prevalence study, a means values one or a comparative one. In this paper we explain some basic topics of statistics and we describe four simple samples of estimation of sample size.
Monaghan, Kieran A.
2016-01-01
Natural ecological variability and analytical design can bias the derived value of a biotic index through the variable influence of indicator body-size, abundance, richness, and ascribed tolerance scores. Descriptive statistics highlight this risk for 26 aquatic indicator systems; detailed analysis is provided for contrasting weighted-average indices applying the example of the BMWP, which has the best supporting data. Differences in body size between taxa from respective tolerance classes is a common feature of indicator systems; in some it represents a trend ranging from comparatively small pollution tolerant to larger intolerant organisms. Under this scenario, the propensity to collect a greater proportion of smaller organisms is associated with negative bias however, positive bias may occur when equipment (e.g. mesh-size) selectively samples larger organisms. Biotic indices are often derived from systems where indicator taxa are unevenly distributed along the gradient of tolerance classes. Such skews in indicator richness can distort index values in the direction of taxonomically rich indicator classes with the subsequent degree of bias related to the treatment of abundance data. The misclassification of indicator taxa causes bias that varies with the magnitude of the misclassification, the relative abundance of misclassified taxa and the treatment of abundance data. These artifacts of assessment design can compromise the ability to monitor biological quality. The statistical treatment of abundance data and the manipulation of indicator assignment and class richness can be used to improve index accuracy. While advances in methods of data collection (i.e. DNA barcoding) may facilitate improvement, the scope to reduce systematic bias is ultimately limited to a strategy of optimal compromise. The shortfall in accuracy must be addressed by statistical pragmatism. At any particular site, the net bias is a probabilistic function of the sample data, resulting in an error variance around an average deviation. Following standardized protocols and assigning precise reference conditions, the error variance of their comparative ratio (test-site:reference) can be measured and used to estimate the accuracy of the resultant assessment. PMID:27392036
NASA Technical Reports Server (NTRS)
Bell, Thomas L.; Kundu, Prasun K.; Kummerow, Christian D.; Einaudi, Franco (Technical Monitor)
2000-01-01
Quantitative use of satellite-derived maps of monthly rainfall requires some measure of the accuracy of the satellite estimates. The rainfall estimate for a given map grid box is subject to both remote-sensing error and, in the case of low-orbiting satellites, sampling error due to the limited number of observations of the grid box provided by the satellite. A simple model of rain behavior predicts that Root-mean-square (RMS) random error in grid-box averages should depend in a simple way on the local average rain rate, and the predicted behavior has been seen in simulations using surface rain-gauge and radar data. This relationship was examined using satellite SSM/I data obtained over the western equatorial Pacific during TOGA COARE. RMS error inferred directly from SSM/I rainfall estimates was found to be larger than predicted from surface data, and to depend less on local rain rate than was predicted. Preliminary examination of TRMM microwave estimates shows better agreement with surface data. A simple method of estimating rms error in satellite rainfall estimates is suggested, based on quantities that can be directly computed from the satellite data.
Estimating fluvial wood discharge from timelapse photography with varying sampling intervals
NASA Astrophysics Data System (ADS)
Anderson, N. K.
2013-12-01
There is recent focus on calculating wood budgets for streams and rivers to help inform management decisions, ecological studies and carbon/nutrient cycling models. Most work has measured in situ wood in temporary storage along stream banks or estimated wood inputs from banks. Little effort has been employed monitoring and quantifying wood in transport during high flows. This paper outlines a procedure for estimating total seasonal wood loads using non-continuous coarse interval sampling and examines differences in estimation between sampling at 1, 5, 10 and 15 minutes. Analysis is performed on wood transport for the Slave River in Northwest Territories, Canada. Relative to the 1 minute dataset, precision decreased by 23%, 46% and 60% for the 5, 10 and 15 minute datasets, respectively. Five and 10 minute sampling intervals provided unbiased equal variance estimates of 1 minute sampling, whereas 15 minute intervals were biased towards underestimation by 6%. Stratifying estimates by day and by discharge increased precision over non-stratification by 4% and 3%, respectively. Not including wood transported during ice break-up, the total minimum wood load estimated at this site is 3300 × 800$ m3 for the 2012 runoff season. The vast majority of the imprecision in total wood volumes came from variance in estimating average volume per log. Comparison of proportions and variance across sample intervals using bootstrap sampling to achieve equal n. Each trial was sampled for n=100, 10,000 times and averaged. All trials were then averaged to obtain an estimate for each sample interval. Dashed lines represent values from the one minute dataset.
NASA Technical Reports Server (NTRS)
Bell, Thomas L.; Kundu, Prasun K.; Einaudi, Franco (Technical Monitor)
2000-01-01
Estimates from TRMM satellite data of monthly total rainfall over an area are subject to substantial sampling errors due to the limited number of visits to the area by the satellite during the month. Quantitative comparisons of TRMM averages with data collected by other satellites and by ground-based systems require some estimate of the size of this sampling error. A method of estimating this sampling error based on the actual statistics of the TRMM observations and on some modeling work has been developed. "Sampling error" in TRMM monthly averages is defined here relative to the monthly total a hypothetical satellite permanently stationed above the area would have reported. "Sampling error" therefore includes contributions from the random and systematic errors introduced by the satellite remote sensing system. As part of our long-term goal of providing error estimates for each grid point accessible to the TRMM instruments, sampling error estimates for TRMM based on rain retrievals from TRMM microwave (TMI) data are compared for different times of the year and different oceanic areas (to minimize changes in the statistics due to algorithmic differences over land and ocean). Changes in sampling error estimates due to changes in rain statistics due 1) to evolution of the official algorithms used to process the data, and 2) differences from other remote sensing systems such as the Defense Meteorological Satellite Program (DMSP) Special Sensor Microwave/Imager (SSM/I), are analyzed.
Error modeling for differential GPS. M.S. Thesis - MIT, 12 May 1995
NASA Technical Reports Server (NTRS)
Blerman, Gregory S.
1995-01-01
Differential Global Positioning System (DGPS) positioning is used to accurately locate a GPS receiver based upon the well-known position of a reference site. In utilizing this technique, several error sources contribute to position inaccuracy. This thesis investigates the error in DGPS operation and attempts to develop a statistical model for the behavior of this error. The model for DGPS error is developed using GPS data collected by Draper Laboratory. The Marquardt method for nonlinear curve-fitting is used to find the parameters of a first order Markov process that models the average errors from the collected data. The results show that a first order Markov process can be used to model the DGPS error as a function of baseline distance and time delay. The model's time correlation constant is 3847.1 seconds (1.07 hours) for the mean square error. The distance correlation constant is 122.8 kilometers. The total process variance for the DGPS model is 3.73 sq meters.
Estimation of Rainfall Sampling Uncertainty: A Comparison of Two Diverse Approaches
NASA Technical Reports Server (NTRS)
Steiner, Matthias; Zhang, Yu; Baeck, Mary Lynn; Wood, Eric F.; Smith, James A.; Bell, Thomas L.; Lau, William K. M. (Technical Monitor)
2002-01-01
The spatial and temporal intermittence of rainfall causes the averages of satellite observations of rain rate to differ from the "true" average rain rate over any given area and time period, even if the satellite observations are perfectly accurate. The difference of satellite averages based on occasional observation by satellite systems and the continuous-time average of rain rate is referred to as sampling error. In this study, rms sampling error estimates are obtained for average rain rates over boxes 100 km, 200 km, and 500 km on a side, for averaging periods of 1 day, 5 days, and 30 days. The study uses a multi-year, merged radar data product provided by Weather Services International Corp. at a resolution of 2 km in space and 15 min in time, over an area of the central U.S. extending from 35N to 45N in latitude and 100W to 80W in longitude. The intervals between satellite observations are assumed to be equal, and similar In size to what present and future satellite systems are able to provide (from 1 h to 12 h). The sampling error estimates are obtained using a resampling method called "resampling by shifts," and are compared to sampling error estimates proposed by Bell based on earlier work by Laughlin. The resampling estimates are found to scale with areal size and time period as the theory predicts. The dependence on average rain rate and time interval between observations is also similar to what the simple theory suggests.
Distribution of the two-sample t-test statistic following blinded sample size re-estimation.
Lu, Kaifeng
2016-05-01
We consider the blinded sample size re-estimation based on the simple one-sample variance estimator at an interim analysis. We characterize the exact distribution of the standard two-sample t-test statistic at the final analysis. We describe a simulation algorithm for the evaluation of the probability of rejecting the null hypothesis at given treatment effect. We compare the blinded sample size re-estimation method with two unblinded methods with respect to the empirical type I error, the empirical power, and the empirical distribution of the standard deviation estimator and final sample size. We characterize the type I error inflation across the range of standardized non-inferiority margin for non-inferiority trials, and derive the adjusted significance level to ensure type I error control for given sample size of the internal pilot study. We show that the adjusted significance level increases as the sample size of the internal pilot study increases. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.
Bayesian inversions of a dynamic vegetation model in four European grassland sites
NASA Astrophysics Data System (ADS)
Minet, J.; Laloy, E.; Tychon, B.; François, L.
2015-01-01
Eddy covariance data from four European grassland sites are used to probabilistically invert the CARAIB dynamic vegetation model (DVM) with ten unknown parameters, using the DREAM(ZS) Markov chain Monte Carlo (MCMC) sampler. We compare model inversions considering both homoscedastic and heteroscedastic eddy covariance residual errors, with variances either fixed a~priori or jointly inferred with the model parameters. Agreements between measured and simulated data during calibration are comparable with previous studies, with root-mean-square error (RMSE) of simulated daily gross primary productivity (GPP), ecosystem respiration (RECO) and evapotranspiration (ET) ranging from 1.73 to 2.19 g C m-2 day-1, 1.04 to 1.56 g C m-2 day-1, and 0.50 to 1.28 mm day-1, respectively. In validation, mismatches between measured and simulated data are larger, but still with Nash-Sutcliffe efficiency scores above 0.5 for three out of the four sites. Although measurement errors associated with eddy covariance data are known to be heteroscedastic, we showed that assuming a classical linear heteroscedastic model of the residual errors in the inversion do not fully remove heteroscedasticity. Since the employed heteroscedastic error model allows for larger deviations between simulated and measured data as the magnitude of the measured data increases, this error model expectedly lead to poorer data fitting compared to inversions considering a constant variance of the residual errors. Furthermore, sampling the residual error variances along with model parameters results in overall similar model parameter posterior distributions as those obtained by fixing these variances beforehand, while slightly improving model performance. Despite the fact that the calibrated model is generally capable of fitting the data within measurement errors, systematic bias in the model simulations are observed. These are likely due to model inadequacies such as shortcomings in the photosynthesis modelling. Besides model behaviour, difference between model parameter posterior distributions among the four grassland sites are also investigated. It is shown that the marginal distributions of the specific leaf area and characteristic mortality time parameters can be explained by site-specific ecophysiological characteristics. Lastly, the possibility of finding a common set of parameters among the four experimental sites is discussed.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Sampson, Andrew; Le Yi; Williamson, Jeffrey F.
2012-02-15
Purpose: To demonstrate potential of correlated sampling Monte Carlo (CMC) simulation to improve the calculation efficiency for permanent seed brachytherapy (PSB) implants without loss of accuracy. Methods: CMC was implemented within an in-house MC code family (PTRAN) and used to compute 3D dose distributions for two patient cases: a clinical PSB postimplant prostate CT imaging study and a simulated post lumpectomy breast PSB implant planned on a screening dedicated breast cone-beam CT patient exam. CMC tallies the dose difference, {Delta}D, between highly correlated histories in homogeneous and heterogeneous geometries. The heterogeneous geometry histories were derived from photon collisions sampled inmore » a geometrically identical but purely homogeneous medium geometry, by altering their particle weights to correct for bias. The prostate case consisted of 78 Model-6711 {sup 125}I seeds. The breast case consisted of 87 Model-200 {sup 103}Pd seeds embedded around a simulated lumpectomy cavity. Systematic and random errors in CMC were unfolded using low-uncertainty uncorrelated MC (UMC) as the benchmark. CMC efficiency gains, relative to UMC, were computed for all voxels, and the mean was classified in regions that received minimum doses greater than 20%, 50%, and 90% of D{sub 90}, as well as for various anatomical regions. Results: Systematic errors in CMC relative to UMC were less than 0.6% for 99% of the voxels and 0.04% for 100% of the voxels for the prostate and breast cases, respectively. For a 1 x 1 x 1 mm{sup 3} dose grid, efficiency gains were realized in all structures with 38.1- and 59.8-fold average gains within the prostate and breast clinical target volumes (CTVs), respectively. Greater than 99% of the voxels within the prostate and breast CTVs experienced an efficiency gain. Additionally, it was shown that efficiency losses were confined to low dose regions while the largest gains were located where little difference exists between the homogeneous and heterogeneous doses. On an AMD 1090T processor, computing times of 38 and 21 sec were required to achieve an average statistical uncertainty of 2% within the prostate (1 x 1 x 1 mm{sup 3}) and breast (0.67 x 0.67 x 0.8 mm{sup 3}) CTVs, respectively. Conclusions: CMC supports an additional average 38-60 fold improvement in average efficiency relative to conventional uncorrelated MC techniques, although some voxels experience no gain or even efficiency losses. However, for the two investigated case studies, the maximum variance within clinically significant structures was always reduced (on average by a factor of 6) in the therapeutic dose range generally. CMC takes only seconds to produce an accurate, high-resolution, low-uncertainly dose distribution for the low-energy PSB implants investigated in this study.« less
Barriers to medication error reporting among hospital nurses.
Rutledge, Dana N; Retrosi, Tina; Ostrowski, Gary
2018-03-01
The study purpose was to report medication error reporting barriers among hospital nurses, and to determine validity and reliability of an existing medication error reporting barriers questionnaire. Hospital medication errors typically occur between ordering of a medication to its receipt by the patient with subsequent staff monitoring. To decrease medication errors, factors surrounding medication errors must be understood; this requires reporting by employees. Under-reporting can compromise patient safety by disabling improvement efforts. This 2017 descriptive study was part of a larger workforce engagement study at a faith-based Magnet ® -accredited community hospital in California (United States). Registered nurses (~1,000) were invited to participate in the online survey via email. Reported here are sample demographics (n = 357) and responses to the 20-item medication error reporting barriers questionnaire. Using factor analysis, four factors that accounted for 67.5% of the variance were extracted. These factors (subscales) were labelled Fear, Cultural Barriers, Lack of Knowledge/Feedback and Practical/Utility Barriers; each demonstrated excellent internal consistency. The medication error reporting barriers questionnaire, originally developed in long-term care, demonstrated good validity and excellent reliability among hospital nurses. Substantial proportions of American hospital nurses (11%-48%) considered specific factors as likely reporting barriers. Average scores on most barrier items were categorised "somewhat unlikely." The highest six included two barriers concerning the time-consuming nature of medication error reporting and four related to nurses' fear of repercussions. Hospitals need to determine the presence of perceived barriers among nurses using questionnaires such as the medication error reporting barriers and work to encourage better reporting. Barriers to medication error reporting make it less likely that nurses will report medication errors, especially errors where patient harm is not apparent or where an error might be hidden. Such under-reporting impedes collection of accurate medication error data and prevents hospitals from changing harmful practices. © 2018 John Wiley & Sons Ltd.
Decorrelation of the true and estimated classifier errors in high-dimensional settings.
Hanczar, Blaise; Hua, Jianping; Dougherty, Edward R
2007-01-01
The aim of many microarray experiments is to build discriminatory diagnosis and prognosis models. Given the huge number of features and the small number of examples, model validity which refers to the precision of error estimation is a critical issue. Previous studies have addressed this issue via the deviation distribution (estimated error minus true error), in particular, the deterioration of cross-validation precision in high-dimensional settings where feature selection is used to mitigate the peaking phenomenon (overfitting). Because classifier design is based upon random samples, both the true and estimated errors are sample-dependent random variables, and one would expect a loss of precision if the estimated and true errors are not well correlated, so that natural questions arise as to the degree of correlation and the manner in which lack of correlation impacts error estimation. We demonstrate the effect of correlation on error precision via a decomposition of the variance of the deviation distribution, observe that the correlation is often severely decreased in high-dimensional settings, and show that the effect of high dimensionality on error estimation tends to result more from its decorrelating effects than from its impact on the variance of the estimated error. We consider the correlation between the true and estimated errors under different experimental conditions using both synthetic and real data, several feature-selection methods, different classification rules, and three error estimators commonly used (leave-one-out cross-validation, k-fold cross-validation, and .632 bootstrap). Moreover, three scenarios are considered: (1) feature selection, (2) known-feature set, and (3) all features. Only the first is of practical interest; however, the other two are needed for comparison purposes. We will observe that the true and estimated errors tend to be much more correlated in the case of a known feature set than with either feature selection or using all features, with the better correlation between the latter two showing no general trend, but differing for different models.
NASA Technical Reports Server (NTRS)
Menard, Richard; Chang, Lang-Ping
1998-01-01
A Kalman filter system designed for the assimilation of limb-sounding observations of stratospheric chemical tracers, which has four tunable covariance parameters, was developed in Part I (Menard et al. 1998) The assimilation results of CH4 observations from the Cryogenic Limb Array Etalon Sounder instrument (CLAES) and the Halogen Observation Experiment instrument (HALOE) on board of the Upper Atmosphere Research Satellite are described in this paper. A robust (chi)(sup 2) criterion, which provides a statistical validation of the forecast and observational error covariances, was used to estimate the tunable variance parameters of the system. In particular, an estimate of the model error variance was obtained. The effect of model error on the forecast error variance became critical after only three days of assimilation of CLAES observations, although it took 14 days of forecast to double the initial error variance. We further found that the model error due to numerical discretization as arising in the standard Kalman filter algorithm, is comparable in size to the physical model error due to wind and transport modeling errors together. Separate assimilations of CLAES and HALOE observations were compared to validate the state estimate away from the observed locations. A wave-breaking event that took place several thousands of kilometers away from the HALOE observation locations was well captured by the Kalman filter due to highly anisotropic forecast error correlations. The forecast error correlation in the assimilation of the CLAES observations was found to have a structure similar to that in pure forecast mode except for smaller length scales. Finally, we have conducted an analysis of the variance and correlation dynamics to determine their relative importance in chemical tracer assimilation problems. Results show that the optimality of a tracer assimilation system depends, for the most part, on having flow-dependent error correlation rather than on evolving the error variance.
NASA Technical Reports Server (NTRS)
Koster, Randal; Walker, Greg; Mahanama, Sarith; Reichle, Rolf
2012-01-01
Continental-scale offline simulations with a land surface model are used to address two important issues in the forecasting of large-scale seasonal streamflow: (i) the extent to which errors in soil moisture initialization degrade streamflow forecasts, and (ii) the extent to which the downscaling of seasonal precipitation forecasts, if it could be done accurately, would improve streamflow forecasts. The reduction in streamflow forecast skill (with forecasted streamflow measured against observations) associated with adding noise to a soil moisture field is found to be, to first order, proportional to the average reduction in the accuracy of the soil moisture field itself. This result has implications for streamflow forecast improvement under satellite-based soil moisture measurement programs. In the second and more idealized ("perfect model") analysis, precipitation downscaling is found to have an impact on large-scale streamflow forecasts only if two conditions are met: (i) evaporation variance is significant relative to the precipitation variance, and (ii) the subgrid spatial variance of precipitation is adequately large. In the large-scale continental region studied (the conterminous United States), these two conditions are met in only a somewhat limited area.
A Stochastic Kinematic Model of Class Averaging in Single-Particle Electron Microscopy
Park, Wooram; Midgett, Charles R.; Madden, Dean R.; Chirikjian, Gregory S.
2011-01-01
Single-particle electron microscopy is an experimental technique that is used to determine the 3D structure of biological macromolecules and the complexes that they form. In general, image processing techniques and reconstruction algorithms are applied to micrographs, which are two-dimensional (2D) images taken by electron microscopes. Each of these planar images can be thought of as a projection of the macromolecular structure of interest from an a priori unknown direction. A class is defined as a collection of projection images with a high degree of similarity, presumably resulting from taking projections along similar directions. In practice, micrographs are very noisy and those in each class are aligned and averaged in order to reduce the background noise. Errors in the alignment process are inevitable due to noise in the electron micrographs. This error results in blurry averaged images. In this paper, we investigate how blurring parameters are related to the properties of the background noise in the case when the alignment is achieved by matching the mass centers and the principal axes of the experimental images. We observe that the background noise in micrographs can be treated as Gaussian. Using the mean and variance of the background Gaussian noise, we derive equations for the mean and variance of translational and rotational misalignments in the class averaging process. This defines a Gaussian probability density on the Euclidean motion group of the plane. Our formulation is validated by convolving the derived blurring function representing the stochasticity of the image alignments with the underlying noiseless projection and comparing with the original blurry image. PMID:21660125
Criterion Predictability: Identifying Differences Between [r-squares
ERIC Educational Resources Information Center
Malgady, Robert G.
1976-01-01
An analysis of variance procedure for testing differences in r-squared, the coefficient of determination, across independent samples is proposed and briefly discussed. The principal advantage of the procedure is to minimize Type I error for follow-up tests of pairwise differences. (Author/JKS)
Gonçalves, Fabio; Treuhaft, Robert; Law, Beverly; ...
2017-01-07
Mapping and monitoring of forest carbon stocks across large areas in the tropics will necessarily rely on remote sensing approaches, which in turn depend on field estimates of biomass for calibration and validation purposes. Here, we used field plot data collected in a tropical moist forest in the central Amazon to gain a better understanding of the uncertainty associated with plot-level biomass estimates obtained specifically for the calibration of remote sensing measurements. In addition to accounting for sources of error that would be normally expected in conventional biomass estimates (e.g., measurement and allometric errors), we examined two sources of uncertaintymore » that are specific to the calibration process and should be taken into account in most remote sensing studies: the error resulting from spatial disagreement between field and remote sensing measurements (i.e., co-location error), and the error introduced when accounting for temporal differences in data acquisition. We found that the overall uncertainty in the field biomass was typically 25% for both secondary and primary forests, but ranged from 16 to 53%. Co-location and temporal errors accounted for a large fraction of the total variance (>65%) and were identified as important targets for reducing uncertainty in studies relating tropical forest biomass to remotely sensed data. Although measurement and allometric errors were relatively unimportant when considered alone, combined they accounted for roughly 30% of the total variance on average and should not be ignored. Lastly, our results suggest that a thorough understanding of the sources of error associated with field-measured plot-level biomass estimates in tropical forests is critical to determine confidence in remote sensing estimates of carbon stocks and fluxes, and to develop strategies for reducing the overall uncertainty of remote sensing approaches.« less
Mumford, Jeanette A.
2017-01-01
Even after thorough preprocessing and a careful time series analysis of functional magnetic resonance imaging (fMRI) data, artifact and other issues can lead to violations of the assumption that the variance is constant across subjects in the group level model. This is especially concerning when modeling a continuous covariate at the group level, as the slope is easily biased by outliers. Various models have been proposed to deal with outliers including models that use the first level variance or that use the group level residual magnitude to differentially weight subjects. The most typically used robust regression, implementing a robust estimator of the regression slope, has been previously studied in the context of fMRI studies and was found to perform well in some scenarios, but a loss of Type I error control can occur for some outlier settings. A second type of robust regression using a heteroscedastic autocorrelation consistent (HAC) estimator, which produces robust slope and variance estimates has been shown to perform well, with better Type I error control, but with large sample sizes (500–1000 subjects). The Type I error control with smaller sample sizes has not been studied in this model and has not been compared to other modeling approaches that handle outliers such as FSL’s Flame 1 and FSL’s outlier de-weighting. Focusing on group level inference with a continuous covariate over a range of sample sizes and degree of heteroscedasticity, which can be driven either by the within- or between-subject variability, both styles of robust regression are compared to ordinary least squares (OLS), FSL’s Flame 1, Flame 1 with outlier de-weighting algorithm and Kendall’s Tau. Additionally, subject omission using the Cook’s Distance measure with OLS and nonparametric inference with the OLS statistic are studied. Pros and cons of these models as well as general strategies for detecting outliers in data and taking precaution to avoid inflated Type I error rates are discussed. PMID:28030782
Zullig, Keith J; Collins, Rani; Ghani, Nadia; Patton, Jon M; Scott Huebner, E; Ajamie, Jean
2014-02-01
The School Climate Measure (SCM) was developed and validated in 2010 in response to a dearth of psychometrically sound school climate instruments. This study sought to further validate the SCM on a large, diverse sample of Arizona public school adolescents (N = 20,953). Four SCM domains (positive student-teacher relationships, academic support, order and discipline, and physical environment) were available for the analysis. Confirmatory factor analysis and structural equation modeling were established to construct validity, and criterion-related validity was assessed via selected Youth Risk Behavior Survey (YRBS) school safety items and self-reported grade (GPA) point average. Analyses confirmed the 4 SCM school climate domains explained approximately 63% of the variance (factor loading range .45-.92). Structural equation models fit the data well χ(2) = 14,325 (df = 293, p < .001), comparative fit index (CFI) = .951, Tuker-Lewis index (TLI) = .952, root mean square error of approximation (RMSEA) = .05). The goodness-of-fit index was .940. Coefficient alphas ranged from .82 to .93. Analyses of variance with post hoc comparisons suggested the SCM domains related in hypothesized directions with the school safety items and GPA. Additional evidence supports the validity and reliability of the SCM. Measures, such as the SCM, can facilitate data-driven decisions and may be incorporated into evidenced-based processes designed to improve student outcomes. © 2014, American School Health Association.
Variance Analysis if Unevenly Spaced Time Series Data
1995-12-01
Daka were subsequently removed from mch simulated data set using typical TWSTFT data patterns to create lwo unevenly spaced sets with average...and techniqw are presented for cowecking errors caused by uneven data spacing in typical TWSTFT daka sets. INTRODUCTION Data points obtained from an...the possible data available. In TWSTFT , the task is less daunting: time transfers are typically measured on Monday, Wednesday, and Friday, so, in a
Simulation Study Using a New Type of Sample Variance
NASA Technical Reports Server (NTRS)
Howe, D. A.; Lainson, K. J.
1996-01-01
We evaluate with simulated data a new type of sample variance for the characterization of frequency stability. The new statistic (referred to as TOTALVAR and its square root TOTALDEV) is a better predictor of long-term frequency variations than the present sample Allan deviation. The statistical model uses the assumption that a time series of phase or frequency differences is wrapped (periodic) with overall frequency difference removed. We find that the variability at long averaging times is reduced considerably for the five models of power-law noise commonly encountered with frequency standards and oscillators.
NASA Astrophysics Data System (ADS)
Musa, Rosliza; Ali, Zalila; Baharum, Adam; Nor, Norlida Mohd
2017-08-01
The linear regression model assumes that all random error components are identically and independently distributed with constant variance. Hence, each data point provides equally precise information about the deterministic part of the total variation. In other words, the standard deviations of the error terms are constant over all values of the predictor variables. When the assumption of constant variance is violated, the ordinary least squares estimator of regression coefficient lost its property of minimum variance in the class of linear and unbiased estimators. Weighted least squares estimation are often used to maximize the efficiency of parameter estimation. A procedure that treats all of the data equally would give less precisely measured points more influence than they should have and would give highly precise points too little influence. Optimizing the weighted fitting criterion to find the parameter estimates allows the weights to determine the contribution of each observation to the final parameter estimates. This study used polynomial model with weighted least squares estimation to investigate paddy production of different paddy lots based on paddy cultivation characteristics and environmental characteristics in the area of Kedah and Perlis. The results indicated that factors affecting paddy production are mixture fertilizer application cycle, average temperature, the squared effect of average rainfall, the squared effect of pest and disease, the interaction between acreage with amount of mixture fertilizer, the interaction between paddy variety and NPK fertilizer application cycle and the interaction between pest and disease and NPK fertilizer application cycle.
NASA Astrophysics Data System (ADS)
Bouhaj, M.; von Estorff, O.; Peiffer, A.
2017-09-01
In the application of Statistical Energy Analysis "SEA" to complex assembled structures, a purely predictive model often exhibits errors. These errors are mainly due to a lack of accurate modelling of the power transmission mechanism described through the Coupling Loss Factors (CLF). Experimental SEA (ESEA) is practically used by the automotive and aerospace industry to verify and update the model or to derive the CLFs for use in an SEA predictive model when analytical estimates cannot be made. This work is particularly motivated by the lack of procedures that allow an estimate to be made of the variance and confidence intervals of the statistical quantities when using the ESEA technique. The aim of this paper is to introduce procedures enabling a statistical description of measured power input, vibration energies and the derived SEA parameters. Particular emphasis is placed on the identification of structural CLFs of complex built-up structures comparing different methods. By adopting a Stochastic Energy Model (SEM), the ensemble average in ESEA is also addressed. For this purpose, expressions are obtained to randomly perturb the energy matrix elements and generate individual samples for the Monte Carlo (MC) technique applied to derive the ensemble averaged CLF. From results of ESEA tests conducted on an aircraft fuselage section, the SEM approach provides a better performance of estimated CLFs compared to classical matrix inversion methods. The expected range of CLF values and the synthesized energy are used as quality criteria of the matrix inversion, allowing to assess critical SEA subsystems, which might require a more refined statistical description of the excitation and the response fields. Moreover, the impact of the variance of the normalized vibration energy on uncertainty of the derived CLFs is outlined.
The Effect of Cluster Sampling Design in Survey Research on the Standard Error Statistic.
ERIC Educational Resources Information Center
Wang, Lin; Fan, Xitao
Standard statistical methods are used to analyze data that is assumed to be collected using a simple random sampling scheme. These methods, however, tend to underestimate variance when the data is collected with a cluster design, which is often found in educational survey research. The purposes of this paper are to demonstrate how a cluster design…
Saviane, Chiara; Silver, R Angus
2006-06-15
Synapses play a crucial role in information processing in the brain. Amplitude fluctuations of synaptic responses can be used to extract information about the mechanisms underlying synaptic transmission and its modulation. In particular, multiple-probability fluctuation analysis can be used to estimate the number of functional release sites, the mean probability of release and the amplitude of the mean quantal response from fits of the relationship between the variance and mean amplitude of postsynaptic responses, recorded at different probabilities. To determine these quantal parameters, calculate their uncertainties and the goodness-of-fit of the model, it is important to weight the contribution of each data point in the fitting procedure. We therefore investigated the errors associated with measuring the variance by determining the best estimators of the variance of the variance and have used simulations of synaptic transmission to test their accuracy and reliability under different experimental conditions. For central synapses, which generally have a low number of release sites, the amplitude distribution of synaptic responses is not normal, thus the use of a theoretical variance of the variance based on the normal assumption is not a good approximation. However, appropriate estimators can be derived for the population and for limited sample sizes using a more general expression that involves higher moments and introducing unbiased estimators based on the h-statistics. Our results are likely to be relevant for various applications of fluctuation analysis when few channels or release sites are present.
NASA Astrophysics Data System (ADS)
De Felice, Matteo; Petitta, Marcello; Ruti, Paolo
2014-05-01
Photovoltaic diffusion is steadily growing on Europe, passing from a capacity of almost 14 GWp in 2011 to 21.5 GWp in 2012 [1]. Having accurate forecast is needed for planning and operational purposes, with the possibility to model and predict solar variability at different time-scales. This study examines the predictability of daily surface solar radiation comparing ECMWF operational forecasts with CM-SAF satellite measurements on the Meteosat (MSG) full disk domain. Operational forecasts used are the IFS system up to 10 days and the System4 seasonal forecast up to three months. Forecast are analysed considering average and variance of errors, showing error maps and average on specific domains with respect to prediction lead times. In all the cases, forecasts are compared with predictions obtained using persistence and state-of-art time-series models. We can observe a wide range of errors, with the performance of forecasts dramatically affected by orography and season. Lower errors are on southern Italy and Spain, with errors on some areas consistently under 10% up to ten days during summer (JJA). Finally, we conclude the study with some insight on how to "translate" the error on solar radiation to error on solar power production using available production data from solar power plants. [1] EurObserver, "Baromètre Photovoltaïque, Le journal des énergies renouvables, April 2012."
Eigenvector method for umbrella sampling enables error analysis
Thiede, Erik H.; Van Koten, Brian; Weare, Jonathan; Dinner, Aaron R.
2016-01-01
Umbrella sampling efficiently yields equilibrium averages that depend on exploring rare states of a model by biasing simulations to windows of coordinate values and then combining the resulting data with physical weighting. Here, we introduce a mathematical framework that casts the step of combining the data as an eigenproblem. The advantage to this approach is that it facilitates error analysis. We discuss how the error scales with the number of windows. Then, we derive a central limit theorem for averages that are obtained from umbrella sampling. The central limit theorem suggests an estimator of the error contributions from individual windows, and we develop a simple and computationally inexpensive procedure for implementing it. We demonstrate this estimator for simulations of the alanine dipeptide and show that it emphasizes low free energy pathways between stable states in comparison to existing approaches for assessing error contributions. Our work suggests the possibility of using the estimator and, more generally, the eigenvector method for umbrella sampling to guide adaptation of the simulation parameters to accelerate convergence. PMID:27586912
Straub, D.E.
1998-01-01
The streamflow-gaging station network in Ohio was evaluated for its effectiveness in providing regional streamflow information. The analysis involved application of the principles of generalized least squares regression between streamflow and climatic and basin characteristics. Regression equations were developed for three flow characteristics: (1) the instantaneous peak flow with a 100-year recurrence interval (P100), (2) the mean annual flow (Qa), and (3) the 7-day, 10-year low flow (7Q10). All active and discontinued gaging stations with 5 or more years of unregulated-streamflow data with respect to each flow characteristic were used to develop the regression equations. The gaging-station network was evaluated for the current (1996) condition of the network and estimated conditions of various network strategies if an additional 5 and 20 years of streamflow data were collected. Any active or discontinued gaging station with (1) less than 5 years of unregulated-streamflow record, (2) previously defined basin and climatic characteristics, and (3) the potential for collection of more unregulated-streamflow record were included in the network strategies involving the additional 5 and 20 years of data. The network analysis involved use of the regression equations, in combination with location, period of record, and cost of operation, to determine the contribution of the data for each gaging station to regional streamflow information. The contribution of each gaging station was based on a cost-weighted reduction of the mean square error (average sampling-error variance) associated with each regional estimating equation. All gaging stations included in the network analysis were then ranked according to their contribution to the regional information for each flow characteristic. The predictive ability of the regression equations developed from the gaging station network could be improved for all three flow characteristics with the collection of additional streamflow data. The addition of new gaging stations to the network would result in an even greater improvement of the accuracy of the regional regression equations. Typically, continued data collection at stations with unregulated streamflow for all flow conditions that had less than 11 years of record with drainage areas smaller than 200 square miles contributed the largest cost-weighted reduction to the average sampling-error variance of the regional estimating equations. The results of the network analyses can be used to prioritize the continued operation of active gaging stations or the reactivation of discontinued gaging stations if the objective is to maximize the regional information content in the streamflow-gaging station network.
Rhodes, Alison M; Tran, Thanh V
2013-02-01
This study examined the equivalence or comparability of the measurement properties of seven selected items measuring posttraumatic growth among self-identified Black (n = 270) and White (n = 707) adult survivors of Hurricane Katrina, using data from the Baseline Survey of the Hurricane Katrina Community Advisory Group Study. Internal consistency reliability was equally good for both groups (Cronbach's alphas = .79), as were correlations between individual scale items and their respective overall scale. Confirmatory factor analysis of a congeneric measurement model of seven selected items of posttraumatic growth showed adequate measures of fit for both groups. The results showed only small variation in magnitude of factor loadings and measurement errors between the two samples. Tests of measurement invariance showed mixed results, but overall indicated that factor loading, error variance, and factor variance were similar between the two samples. These seven selected items can be useful for future large-scale surveys of posttraumatic growth.
NASA Astrophysics Data System (ADS)
Leite, Argentina; Paula Rocha, Ana; Eduarda Silva, Maria
2013-06-01
Heart Rate Variability (HRV) series exhibit long memory and time-varying conditional variance. This work considers the Fractionally Integrated AutoRegressive Moving Average (ARFIMA) models with Generalized AutoRegressive Conditional Heteroscedastic (GARCH) errors. ARFIMA-GARCH models may be used to capture and remove long memory and estimate the conditional volatility in 24 h HRV recordings. The ARFIMA-GARCH approach is applied to fifteen long term HRV series available at Physionet, leading to the discrimination among normal individuals, heart failure patients, and patients with atrial fibrillation.
Empirical Bayes estimation of undercount in the decennial census.
Cressie, N
1989-12-01
Empirical Bayes methods are used to estimate the extent of the undercount at the local level in the 1980 U.S. census. "Grouping of like subareas from areas such as states, counties, and so on into strata is a useful way of reducing the variance of undercount estimators. By modeling the subareas within a stratum to have a common mean and variances inversely proportional to their census counts, and by taking into account sampling of the areas (e.g., by dual-system estimation), empirical Bayes estimators that compromise between the (weighted) stratum average and the sample value can be constructed. The amount of compromise is shown to depend on the relative importance of stratum variance to sampling variance. These estimators are evaluated at the state level (51 states, including Washington, D.C.) and stratified on race/ethnicity (3 strata) using data from the 1980 postenumeration survey (PEP 3-8, for the noninstitutional population)." excerpt
Do stochastic inhomogeneities affect dark-energy precision measurements?
Ben-Dayan, I; Gasperini, M; Marozzi, G; Nugier, F; Veneziano, G
2013-01-11
The effect of a stochastic background of cosmological perturbations on the luminosity-redshift relation is computed to second order through a recently proposed covariant and gauge-invariant light-cone averaging procedure. The resulting expressions are free from both ultraviolet and infrared divergences, implying that such perturbations cannot mimic a sizable fraction of dark energy. Different averages are estimated and depend on the particular function of the luminosity distance being averaged. The energy flux being minimally affected by perturbations at large z is proposed as the best choice for precision estimates of dark-energy parameters. Nonetheless, its irreducible (stochastic) variance induces statistical errors on Ω(Λ)(z) typically lying in the few-percent range.
Poston, Brach; Van Gemmert, Arend W.A.; Sharma, Siddharth; Chakrabarti, Somesh; Zavaremi, Shahrzad H.; Stelmach, George
2013-01-01
The minimum variance theory proposes that motor commands are corrupted by signal-dependent noise and smooth trajectories with low noise levels are selected to minimize endpoint error and endpoint variability. The purpose of the study was to determine the contribution of trajectory smoothness to the endpoint accuracy and endpoint variability of rapid multi-joint arm movements. Young and older adults performed arm movements (4 blocks of 25 trials) as fast and as accurately as possible to a target with the right (dominant) arm. Endpoint accuracy and endpoint variability along with trajectory smoothness and error were quantified for each block of trials. Endpoint error and endpoint variance were greater in older adults compared with young adults, but decreased at a similar rate with practice for the two age groups. The greater endpoint error and endpoint variance exhibited by older adults were primarily due to impairments in movement extent control and not movement direction control. The normalized jerk was similar for the two age groups, but was not strongly associated with endpoint error or endpoint variance for either group. However, endpoint variance was strongly associated with endpoint error for both the young and older adults. Finally, trajectory error was similar for both groups and was weakly associated with endpoint error for the older adults. The findings are not consistent with the predictions of the minimum variance theory, but support and extend previous observations that movement trajectories and endpoints are planned independently. PMID:23584101
Image correlation and sampling study
NASA Technical Reports Server (NTRS)
Popp, D. J.; Mccormack, D. S.; Sedwick, J. L.
1972-01-01
The development of analytical approaches for solving image correlation and image sampling of multispectral data is discussed. Relevant multispectral image statistics which are applicable to image correlation and sampling are identified. The general image statistics include intensity mean, variance, amplitude histogram, power spectral density function, and autocorrelation function. The translation problem associated with digital image registration and the analytical means for comparing commonly used correlation techniques are considered. General expressions for determining the reconstruction error for specific image sampling strategies are developed.
New method of extracting information of arterial oxygen saturation based on ∑ | 𝚫 |
NASA Astrophysics Data System (ADS)
Dai, Wenting; Lin, Ling; Li, Gang
2017-04-01
Noninvasive detection of oxygen saturation with near-infrared spectroscopy has been widely used in clinics. In order to further enhance its detection precision and reliability, this paper proposes a method of time domain absolute difference summation (∑|Δ|) based on a dynamic spectrum. In this method, the ratio of absolute differences between intervals of two differential sampling points at the same moment on logarithm photoplethysmography signals of red and infrared light was obtained in turn, and then they obtained a ratio sequence which was screened with a statistical method. Finally, use the summation of the screened ratio sequence as the oxygen saturation coefficient Q. We collected 120 reference samples of SpO2 and then compared the result of two methods, which are ∑|Δ| and peak-peak. Average root-mean-square errors of the two methods were 3.02% and 6.80%, respectively, in the 20 cases which were selected randomly. In addition, the average variance of Q of the 10 samples, which were obtained by the new method, reduced to 22.77% of that obtained by the peak-peak method. Comparing with the commercial product, the new method makes the results more accurate. Theoretical and experimental analysis indicates that the application of the ∑|Δ| method could enhance the precision and reliability of oxygen saturation detection in real time.
New method of extracting information of arterial oxygen saturation based on ∑|𝚫 |
NASA Astrophysics Data System (ADS)
Wenting, Dai; Ling, Lin; Gang, Li
2017-04-01
Noninvasive detection of oxygen saturation with near-infrared spectroscopy has been widely used in clinics. In order to further enhance its detection precision and reliability, this paper proposes a method of time domain absolute difference summation (∑|Δ|) based on a dynamic spectrum. In this method, the ratio of absolute differences between intervals of two differential sampling points at the same moment on logarithm photoplethysmography signals of red and infrared light was obtained in turn, and then they obtained a ratio sequence which was screened with a statistical method. Finally, use the summation of the screened ratio sequence as the oxygen saturation coefficient Q. We collected 120 reference samples of SpO2 and then compared the result of two methods, which are ∑|Δ| and peak-peak. Average root-mean-square errors of the two methods were 3.02% and 6.80%, respectively, in the 20 cases which were selected randomly. In addition, the average variance of Q of the 10 samples, which were obtained by the new method, reduced to 22.77% of that obtained by the peak-peak method. Comparing with the commercial product, the new method makes the results more accurate. Theoretical and experimental analysis indicates that the application of the ∑|Δ| method could enhance the precision and reliability of oxygen saturation detection in real time.
Bayesian Model Averaging of Artificial Intelligence Models for Hydraulic Conductivity Estimation
NASA Astrophysics Data System (ADS)
Nadiri, A.; Chitsazan, N.; Tsai, F. T.; Asghari Moghaddam, A.
2012-12-01
This research presents a Bayesian artificial intelligence model averaging (BAIMA) method that incorporates multiple artificial intelligence (AI) models to estimate hydraulic conductivity and evaluate estimation uncertainties. Uncertainty in the AI model outputs stems from error in model input as well as non-uniqueness in selecting different AI methods. Using one single AI model tends to bias the estimation and underestimate uncertainty. BAIMA employs Bayesian model averaging (BMA) technique to address the issue of using one single AI model for estimation. BAIMA estimates hydraulic conductivity by averaging the outputs of AI models according to their model weights. In this study, the model weights were determined using the Bayesian information criterion (BIC) that follows the parsimony principle. BAIMA calculates the within-model variances to account for uncertainty propagation from input data to AI model output. Between-model variances are evaluated to account for uncertainty due to model non-uniqueness. We employed Takagi-Sugeno fuzzy logic (TS-FL), artificial neural network (ANN) and neurofuzzy (NF) to estimate hydraulic conductivity for the Tasuj plain aquifer, Iran. BAIMA combined three AI models and produced better fitting than individual models. While NF was expected to be the best AI model owing to its utilization of both TS-FL and ANN models, the NF model is nearly discarded by the parsimony principle. The TS-FL model and the ANN model showed equal importance although their hydraulic conductivity estimates were quite different. This resulted in significant between-model variances that are normally ignored by using one AI model.
Comment on Hoffman and Rovine (2007): SPSS MIXED can estimate models with heterogeneous variances.
Weaver, Bruce; Black, Ryan A
2015-06-01
Hoffman and Rovine (Behavior Research Methods, 39:101-117, 2007) have provided a very nice overview of how multilevel models can be useful to experimental psychologists. They included two illustrative examples and provided both SAS and SPSS commands for estimating the models they reported. However, upon examining the SPSS syntax for the models reported in their Table 3, we found no syntax for models 2B and 3B, both of which have heterogeneous error variances. Instead, there is syntax that estimates similar models with homogeneous error variances and a comment stating that SPSS does not allow heterogeneous errors. But that is not correct. We provide SPSS MIXED commands to estimate models 2B and 3B with heterogeneous error variances and obtain results nearly identical to those reported by Hoffman and Rovine in their Table 3. Therefore, contrary to the comment in Hoffman and Rovine's syntax file, SPSS MIXED can estimate models with heterogeneous error variances.
Gains in accuracy from averaging ratings of abnormality
NASA Astrophysics Data System (ADS)
Swensson, Richard G.; King, Jill L.; Gur, David; Good, Walter F.
1999-05-01
Six radiologists used continuous scales to rate 529 chest-film cases for likelihood of five separate types of abnormalities (interstitial disease, nodules, pneumothorax, alveolar infiltrates and rib fractures) in each of six replicated readings, yielding 36 separate ratings of each case for the five abnormalities. Analyses for each type of abnormality estimated the relative gains in accuracy (area below the ROC curve) obtained by averaging the case-ratings across: (1) six independent replications by each reader (30% gain), (2) six different readers within each replication (39% gain) or (3) all 36 readings (58% gain). Although accuracy differed among both readers and abnormalities, ROC curves for the median ratings showed similar relative gains in accuracy. From a latent-variable model for these gains, we estimate that about 51% of a reader's total decision variance consisted of random (within-reader) errors that were uncorrelated between replications, another 14% came from that reader's consistent (but idiosyncratic) responses to different cases, and only about 35% could be attributed to systematic variations among the sampled cases that were consistent across different readers.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Gonçalves, Fabio; Treuhaft, Robert; Law, Beverly
Mapping and monitoring of forest carbon stocks across large areas in the tropics will necessarily rely on remote sensing approaches, which in turn depend on field estimates of biomass for calibration and validation purposes. Here, we used field plot data collected in a tropical moist forest in the central Amazon to gain a better understanding of the uncertainty associated with plot-level biomass estimates obtained specifically for the calibration of remote sensing measurements. In addition to accounting for sources of error that would be normally expected in conventional biomass estimates (e.g., measurement and allometric errors), we examined two sources of uncertaintymore » that are specific to the calibration process and should be taken into account in most remote sensing studies: the error resulting from spatial disagreement between field and remote sensing measurements (i.e., co-location error), and the error introduced when accounting for temporal differences in data acquisition. We found that the overall uncertainty in the field biomass was typically 25% for both secondary and primary forests, but ranged from 16 to 53%. Co-location and temporal errors accounted for a large fraction of the total variance (>65%) and were identified as important targets for reducing uncertainty in studies relating tropical forest biomass to remotely sensed data. Although measurement and allometric errors were relatively unimportant when considered alone, combined they accounted for roughly 30% of the total variance on average and should not be ignored. Lastly, our results suggest that a thorough understanding of the sources of error associated with field-measured plot-level biomass estimates in tropical forests is critical to determine confidence in remote sensing estimates of carbon stocks and fluxes, and to develop strategies for reducing the overall uncertainty of remote sensing approaches.« less
Adaptive use of research aircraft data sets for hurricane forecasts
NASA Astrophysics Data System (ADS)
Biswas, M. K.; Krishnamurti, T. N.
2008-02-01
This study uses an adaptive observational strategy for hurricane forecasting. It shows the impacts of Lidar Atmospheric Sensing Experiment (LASE) and dropsonde data sets from Convection and Moisture Experiment (CAMEX) field campaigns on hurricane track and intensity forecasts. The following cases are used in this study: Bonnie, Danielle and Georges of 1998 and Erin, Gabrielle and Humberto of 2001. A single model run for each storm is carried out using the Florida State University Global Spectral Model (FSUGSM) with the European Center for Medium Range Weather Forecasts (ECMWF) analysis as initial conditions, in addition to 50 other model runs where the analysis is randomly perturbed for each storm. The centers of maximum variance of the DLM heights are located from the forecast error variance fields at the 84-hr forecast. Back correlations are then performed using the centers of these maximum variances and the fields at the 36-hr forecast. The regions having the highest correlations in the vicinity of the hurricanes are indicative of regions from where the error growth emanates and suggests the need for additional observations. Data sets are next assimilated in those areas that contain high correlations. Forecasts are computed using the new initial conditions for the storm cases, and track and intensity skills are then examined with respect to the control forecast. The adaptive strategy is capable of identifying sensitive areas where additional observations can help in reducing the hurricane track forecast errors. A reduction of position error by approximately 52% for day 3 of forecast (averaged over 7 storm cases) over the control runs is observed. The intensity forecast shows only a slight positive impact due to the model’s coarse resolution.
Effect of correlated observation error on parameters, predictions, and uncertainty
Tiedeman, Claire; Green, Christopher T.
2013-01-01
Correlations among observation errors are typically omitted when calculating observation weights for model calibration by inverse methods. We explore the effects of omitting these correlations on estimates of parameters, predictions, and uncertainties. First, we develop a new analytical expression for the difference in parameter variance estimated with and without error correlations for a simple one-parameter two-observation inverse model. Results indicate that omitting error correlations from both the weight matrix and the variance calculation can either increase or decrease the parameter variance, depending on the values of error correlation (ρ) and the ratio of dimensionless scaled sensitivities (rdss). For small ρ, the difference in variance is always small, but for large ρ, the difference varies widely depending on the sign and magnitude of rdss. Next, we consider a groundwater reactive transport model of denitrification with four parameters and correlated geochemical observation errors that are computed by an error-propagation approach that is new for hydrogeologic studies. We compare parameter estimates, predictions, and uncertainties obtained with and without the error correlations. Omitting the correlations modestly to substantially changes parameter estimates, and causes both increases and decreases of parameter variances, consistent with the analytical expression. Differences in predictions for the models calibrated with and without error correlations can be greater than parameter differences when both are considered relative to their respective confidence intervals. These results indicate that including observation error correlations in weighting for nonlinear regression can have important effects on parameter estimates, predictions, and their respective uncertainties.
The efficacy of a novel mobile phone application for goldmann ptosis visual field interpretation.
Maamari, Robi N; D'Ambrosio, Michael V; Joseph, Jeffrey M; Tao, Jeremiah P
2014-01-01
To evaluate the efficacy of a novel mobile phone application that calculates superior visual field defects on Goldmann visual field charts. Experimental study in which the mobile phone application and 14 oculoplastic surgeons interpreted the superior visual field defect in 10 Goldmann charts. Percent error of the mobile phone application and the oculoplastic surgeons' estimates were calculated compared with computer software computation of the actual defects. Precision and time efficiency of the application were evaluated by processing the same Goldmann visual field chart 10 repeated times. The mobile phone application was associated with a mean percent error of 1.98% (95% confidence interval[CI], 0.87%-3.10%) in superior visual field defect calculation. The average mean percent error of the oculoplastic surgeons' visual estimates was 19.75% (95% CI, 14.39%-25.11%). Oculoplastic surgeons, on average, underestimated the defect in all 10 Goldmann charts. There was high interobserver variance among oculoplastic surgeons. The percent error of the 10 repeated measurements on a single chart was 0.93% (95% CI, 0.40%-1.46%). The average time to process 1 chart was 12.9 seconds (95% CI, 10.9-15.0 seconds). The mobile phone application was highly accurate, precise, and time-efficient in calculating the percent superior visual field defect using Goldmann charts. Oculoplastic surgeon visual interpretations were highly inaccurate, highly variable, and usually underestimated the field vision loss.
Age-related variation in genetic control of height growth in Douglas-fir.
Namkoong, G; Usanis, R A; Silen, R R
1972-01-01
The development of genetic variances in height growth of Douglas-fir over a 53-year period is analyzed and found to fall into three periods. In the juvenile period, variances in environmental error increase logarithmically, genetic variance within populations exists at moderate levels, and variance among populations is low but increasing. In the early reproductive period, the response to environmental sources of error variance is restricted, genetic variance within populations disappears, and populational differences strongly emerge but do not increase as expected. In the later period, environmental error again increases rapidly, but genetic variance within populations does not reappear and population differences are maintained at about the same level as established in the early reproductive period. The change between the juvenile and early reproductive periods is perhaps associated with the onset of ecological dominance and significant allocations of energy to reproduction.
Visentin, G; Penasa, M; Gottardo, P; Cassandro, M; De Marchi, M
2016-10-01
Milk minerals and coagulation properties are important for both consumers and processors, and they can aid in increasing milk added value. However, large-scale monitoring of these traits is hampered by expensive and time-consuming reference analyses. The objective of the present study was to develop prediction models for major mineral contents (Ca, K, Mg, Na, and P) and milk coagulation properties (MCP: rennet coagulation time, curd-firming time, and curd firmness) using mid-infrared spectroscopy. Individual milk samples (n=923) of Holstein-Friesian, Brown Swiss, Alpine Grey, and Simmental cows were collected from single-breed herds between January and December 2014. Reference analysis for the determination of both mineral contents and MCP was undertaken with standardized methods. For each milk sample, the mid-infrared spectrum in the range from 900 to 5,000cm(-1) was stored. Prediction models were calibrated using partial least squares regression coupled with a wavenumber selection technique called uninformative variable elimination, to improve model accuracy, and validated both internally and externally. The average reduction of wavenumbers used in partial least squares regression was 80%, which was accompanied by an average increment of 20% of the explained variance in external validation. The proportion of explained variance in external validation was about 70% for P, K, Ca, and Mg, and it was lower (40%) for Na. Milk coagulation properties prediction models explained between 54% (rennet coagulation time) and 56% (curd-firming time) of the total variance in external validation. The ratio of standard deviation of each trait to the respective root mean square error of prediction, which is an indicator of the predictive ability of an equation, suggested that the developed models might be effective for screening and collection of milk minerals and coagulation properties at the population level. Although prediction equations were not accurate enough to be proposed for analytic purposes, mid-infrared spectroscopy predictions could be evaluated as phenotypic information to genetically improve milk minerals and MCP on a large scale. Copyright © 2016 American Dairy Science Association. Published by Elsevier Inc. All rights reserved.
NASA Astrophysics Data System (ADS)
Phillips, Thomas J.; Gates, W. Lawrence; Arpe, Klaus
1992-12-01
The effects of sampling frequency on the first- and second-moment statistics of selected European Centre for Medium-Range Weather Forecasts (ECMWF) model variables are investigated in a simulation of "perpetual July" with a diurnal cycle included and with surface and atmospheric fields saved at hourly intervals. The shortest characteristic time scales (as determined by the e-folding time of lagged autocorrelation functions) are those of ground heat fluxes and temperatures, precipitation and runoff, convective processes, cloud properties, and atmospheric vertical motion, while the longest time scales are exhibited by soil temperature and moisture, surface pressure, and atmospheric specific humidity, temperature, and wind. The time scales of surface heat and momentum fluxes and of convective processes are substantially shorter over land than over oceans. An appropriate sampling frequency for each model variable is obtained by comparing the estimates of first- and second-moment statistics determined at intervals ranging from 2 to 24 hours with the "best" estimates obtained from hourly sampling. Relatively accurate estimation of first- and second-moment climate statistics (10% errors in means, 20% errors in variances) can be achieved by sampling a model variable at intervals that usually are longer than the bandwidth of its time series but that often are shorter than its characteristic time scale. For the surface variables, sampling at intervals that are nonintegral divisors of a 24-hour day yields relatively more accurate time-mean statistics because of a reduction in errors associated with aliasing of the diurnal cycle and higher-frequency harmonics. The superior estimates of first-moment statistics are accompanied by inferior estimates of the variance of the daily means due to the presence of systematic biases, but these probably can be avoided by defining a different measure of low-frequency variability. Estimates of the intradiurnal variance of accumulated precipitation and surface runoff also are strongly impacted by the length of the storage interval. In light of these results, several alternative strategies for storage of the EMWF model variables are recommended.
Estimating Model Prediction Error: Should You Treat Predictions as Fixed or Random?
NASA Technical Reports Server (NTRS)
Wallach, Daniel; Thorburn, Peter; Asseng, Senthold; Challinor, Andrew J.; Ewert, Frank; Jones, James W.; Rotter, Reimund; Ruane, Alexander
2016-01-01
Crop models are important tools for impact assessment of climate change, as well as for exploring management options under current climate. It is essential to evaluate the uncertainty associated with predictions of these models. We compare two criteria of prediction error; MSEP fixed, which evaluates mean squared error of prediction for a model with fixed structure, parameters and inputs, and MSEP uncertain( X), which evaluates mean squared error averaged over the distributions of model structure, inputs and parameters. Comparison of model outputs with data can be used to estimate the former. The latter has a squared bias term, which can be estimated using hindcasts, and a model variance term, which can be estimated from a simulation experiment. The separate contributions to MSEP uncertain (X) can be estimated using a random effects ANOVA. It is argued that MSEP uncertain (X) is the more informative uncertainty criterion, because it is specific to each prediction situation.
Estimation and correction of visibility bias in aerial surveys of wintering ducks
Pearse, A.T.; Gerard, P.D.; Dinsmore, S.J.; Kaminski, R.M.; Reinecke, K.J.
2008-01-01
Incomplete detection of all individuals leading to negative bias in abundance estimates is a pervasive source of error in aerial surveys of wildlife, and correcting that bias is a critical step in improving surveys. We conducted experiments using duck decoys as surrogates for live ducks to estimate bias associated with surveys of wintering ducks in Mississippi, USA. We found detection of decoy groups was related to wetland cover type (open vs. forested), group size (1?100 decoys), and interaction of these variables. Observers who detected decoy groups reported counts that averaged 78% of the decoys actually present, and this counting bias was not influenced by either covariate cited above. We integrated this sightability model into estimation procedures for our sample surveys with weight adjustments derived from probabilities of group detection (estimated by logistic regression) and count bias. To estimate variances of abundance estimates, we used bootstrap resampling of transects included in aerial surveys and data from the bias-correction experiment. When we implemented bias correction procedures on data from a field survey conducted in January 2004, we found bias-corrected estimates of abundance increased 36?42%, and associated standard errors increased 38?55%, depending on species or group estimated. We deemed our method successful for integrating correction of visibility bias in an existing sample survey design for wintering ducks in Mississippi, and we believe this procedure could be implemented in a variety of sampling problems for other locations and species.
Improved uncertainty quantification in nondestructive assay for nonproliferation
DOE Office of Scientific and Technical Information (OSTI.GOV)
Burr, Tom; Croft, Stephen; Jarman, Ken
2016-12-01
This paper illustrates methods to improve uncertainty quantification (UQ) for non-destructive assay (NDA) measurements used in nuclear nonproliferation. First, it is shown that current bottom-up UQ applied to calibration data is not always adequate, for three main reasons: (1) Because there are errors in both the predictors and the response, calibration involves a ratio of random quantities, and calibration data sets in NDA usually consist of only a modest number of samples (3–10); therefore, asymptotic approximations involving quantities needed for UQ such as means and variances are often not sufficiently accurate; (2) Common practice overlooks that calibration implies a partitioningmore » of total error into random and systematic error, and (3) In many NDA applications, test items exhibit non-negligible departures in physical properties from calibration items, so model-based adjustments are used, but item-specific bias remains in some data. Therefore, improved bottom-up UQ using calibration data should predict the typical magnitude of item-specific bias, and the suggestion is to do so by including sources of item-specific bias in synthetic calibration data that is generated using a combination of modeling and real calibration data. Second, for measurements of the same nuclear material item by both the facility operator and international inspectors, current empirical (top-down) UQ is described for estimating operator and inspector systematic and random error variance components. A Bayesian alternative is introduced that easily accommodates constraints on variance components, and is more robust than current top-down methods to the underlying measurement error distributions.« less
A multiple-objective optimal exploration strategy
Christakos, G.; Olea, R.A.
1988-01-01
Exploration for natural resources is accomplished through partial sampling of extensive domains. Such imperfect knowledge is subject to sampling error. Complex systems of equations resulting from modelling based on the theory of correlated random fields are reduced to simple analytical expressions providing global indices of estimation variance. The indices are utilized by multiple objective decision criteria to find the best sampling strategies. The approach is not limited by geometric nature of the sampling, covers a wide range in spatial continuity and leads to a step-by-step procedure. ?? 1988.
Evaluation of three lidar scanning strategies for turbulence measurements
NASA Astrophysics Data System (ADS)
Newman, J. F.; Klein, P. M.; Wharton, S.; Sathe, A.; Bonin, T. A.; Chilson, P. B.; Muschinski, A.
2015-11-01
Several errors occur when a traditional Doppler-beam swinging (DBS) or velocity-azimuth display (VAD) strategy is used to measure turbulence with a lidar. To mitigate some of these errors, a scanning strategy was recently developed which employs six beam positions to independently estimate the u, v, and w velocity variances and covariances. In order to assess the ability of these different scanning techniques to measure turbulence, a Halo scanning lidar, WindCube v2 pulsed lidar and ZephIR continuous wave lidar were deployed at field sites in Oklahoma and Colorado with collocated sonic anemometers. Results indicate that the six-beam strategy mitigates some of the errors caused by VAD and DBS scans, but the strategy is strongly affected by errors in the variance measured at the different beam positions. The ZephIR and WindCube lidars overestimated horizontal variance values by over 60 % under unstable conditions as a result of variance contamination, where additional variance components contaminate the true value of the variance. A correction method was developed for the WindCube lidar that uses variance calculated from the vertical beam position to reduce variance contamination in the u and v variance components. The correction method reduced WindCube variance estimates by over 20 % at both the Oklahoma and Colorado sites under unstable conditions, when variance contamination is largest. This correction method can be easily applied to other lidars that contain a vertical beam position and is a promising method for accurately estimating turbulence with commercially available lidars.
Evaluation of three lidar scanning strategies for turbulence measurements
NASA Astrophysics Data System (ADS)
Newman, Jennifer F.; Klein, Petra M.; Wharton, Sonia; Sathe, Ameya; Bonin, Timothy A.; Chilson, Phillip B.; Muschinski, Andreas
2016-05-01
Several errors occur when a traditional Doppler beam swinging (DBS) or velocity-azimuth display (VAD) strategy is used to measure turbulence with a lidar. To mitigate some of these errors, a scanning strategy was recently developed which employs six beam positions to independently estimate the u, v, and w velocity variances and covariances. In order to assess the ability of these different scanning techniques to measure turbulence, a Halo scanning lidar, WindCube v2 pulsed lidar, and ZephIR continuous wave lidar were deployed at field sites in Oklahoma and Colorado with collocated sonic anemometers.Results indicate that the six-beam strategy mitigates some of the errors caused by VAD and DBS scans, but the strategy is strongly affected by errors in the variance measured at the different beam positions. The ZephIR and WindCube lidars overestimated horizontal variance values by over 60 % under unstable conditions as a result of variance contamination, where additional variance components contaminate the true value of the variance. A correction method was developed for the WindCube lidar that uses variance calculated from the vertical beam position to reduce variance contamination in the u and v variance components. The correction method reduced WindCube variance estimates by over 20 % at both the Oklahoma and Colorado sites under unstable conditions, when variance contamination is largest. This correction method can be easily applied to other lidars that contain a vertical beam position and is a promising method for accurately estimating turbulence with commercially available lidars.
[Theory, method and application of method R on estimation of (co)variance components].
Liu, Wen-Zhong
2004-07-01
Theory, method and application of Method R on estimation of (co)variance components were reviewed in order to make the method be reasonably used. Estimation requires R values,which are regressions of predicted random effects that are calculated using complete dataset on predicted random effects that are calculated using random subsets of the same data. By using multivariate iteration algorithm based on a transformation matrix,and combining with the preconditioned conjugate gradient to solve the mixed model equations, the computation efficiency of Method R is much improved. Method R is computationally inexpensive,and the sampling errors and approximate credible intervals of estimates can be obtained. Disadvantages of Method R include a larger sampling variance than other methods for the same data,and biased estimates in small datasets. As an alternative method, Method R can be used in larger datasets. It is necessary to study its theoretical properties and broaden its application range further.
Re-estimating sample size in cluster randomised trials with active recruitment within clusters.
van Schie, S; Moerbeek, M
2014-08-30
Often only a limited number of clusters can be obtained in cluster randomised trials, although many potential participants can be recruited within each cluster. Thus, active recruitment is feasible within the clusters. To obtain an efficient sample size in a cluster randomised trial, the cluster level and individual level variance should be known before the study starts, but this is often not the case. We suggest using an internal pilot study design to address this problem of unknown variances. A pilot can be useful to re-estimate the variances and re-calculate the sample size during the trial. Using simulated data, it is shown that an initially low or high power can be adjusted using an internal pilot with the type I error rate remaining within an acceptable range. The intracluster correlation coefficient can be re-estimated with more precision, which has a positive effect on the sample size. We conclude that an internal pilot study design may be used if active recruitment is feasible within a limited number of clusters. Copyright © 2014 John Wiley & Sons, Ltd.
NASA Technical Reports Server (NTRS)
Natarajan, Suresh; Gardner, C. S.
1987-01-01
Receiver timing synchronization of an optical Pulse-Position Modulation (PPM) communication system can be achieved using a phased-locked loop (PLL), provided the photodetector output is suitably processed. The magnitude of the PLL phase error is a good indicator of the timing error at the receiver decoder. The statistics of the phase error are investigated while varying several key system parameters such as PPM order, signal and background strengths, and PPL bandwidth. A practical optical communication system utilizing a laser diode transmitter and an avalanche photodiode in the receiver is described, and the sampled phase error data are presented. A linear regression analysis is applied to the data to obtain estimates of the relational constants involving the phase error variance and incident signal power.
Determining Optimal Location and Numbers of Sample Transects for Characterization of UXO Sites
DOE Office of Scientific and Technical Information (OSTI.GOV)
BILISOLY, ROGER L.; MCKENNA, SEAN A.
2003-01-01
Previous work on sample design has been focused on constructing designs for samples taken at point locations. Significantly less work has been done on sample design for data collected along transects. A review of approaches to point and transect sampling design shows that transects can be considered as a sequential set of point samples. Any two sampling designs can be compared through using each one to predict the value of the quantity being measured on a fixed reference grid. The quality of a design is quantified in two ways: computing either the sum or the product of the eigenvalues ofmore » the variance matrix of the prediction error. An important aspect of this analysis is that the reduction of the mean prediction error variance (MPEV) can be calculated for any proposed sample design, including one with straight and/or meandering transects, prior to taking those samples. This reduction in variance can be used as a ''stopping rule'' to determine when enough transect sampling has been completed on the site. Two approaches for the optimization of the transect locations are presented. The first minimizes the sum of the eigenvalues of the predictive error, and the second minimizes the product of these eigenvalues. Simulated annealing is used to identify transect locations that meet either of these objectives. This algorithm is applied to a hypothetical site to determine the optimal locations of two iterations of meandering transects given a previously existing straight transect. The MPEV calculation is also used on both a hypothetical site and on data collected at the Isleta Pueblo to evaluate its potential as a stopping rule. Results show that three or four rounds of systematic sampling with straight parallel transects covering 30 percent or less of the site, can reduce the initial MPEV by as much as 90 percent. The amount of reduction in MPEV can be used as a stopping rule, but the relationship between MPEV and the results of excavation versus no-further-action decisions is site specific and cannot be calculated prior to the sampling. It may be advantageous to use the reduction in MPEV as a stopping rule for systematic sampling across the site that can then be followed by focused sampling in areas identified has having UXO during the systematic sampling. The techniques presented here provide answers to the questions of ''Where to sample?'' and ''When to stop?'' and are capable of running in near real time to support iterative site characterization campaigns.« less
Genetic and Environmental Contributions to Educational Attainment in Australia.
ERIC Educational Resources Information Center
Miller, Paul; Mulvey, Charles; Martin, Nick
2001-01-01
Data from a large sample of Australian twins indicate that 50 to 65 percent of variance in educational attainments can be attributed to genetic endowments. Only about 25 to 40 percent may be due to environmental factors, depending on adjustments for measurement error and assortative mating. (Contains 51 references.) (MLH)
He, Jianbo; Li, Jijie; Huang, Zhongwen; Zhao, Tuanjie; Xing, Guangnan; Gai, Junyi; Guan, Rongzhan
2015-01-01
Experimental error control is very important in quantitative trait locus (QTL) mapping. Although numerous statistical methods have been developed for QTL mapping, a QTL detection model based on an appropriate experimental design that emphasizes error control has not been developed. Lattice design is very suitable for experiments with large sample sizes, which is usually required for accurate mapping of quantitative traits. However, the lack of a QTL mapping method based on lattice design dictates that the arithmetic mean or adjusted mean of each line of observations in the lattice design had to be used as a response variable, resulting in low QTL detection power. As an improvement, we developed a QTL mapping method termed composite interval mapping based on lattice design (CIMLD). In the lattice design, experimental errors are decomposed into random errors and block-within-replication errors. Four levels of block-within-replication errors were simulated to show the power of QTL detection under different error controls. The simulation results showed that the arithmetic mean method, which is equivalent to a method under random complete block design (RCBD), was very sensitive to the size of the block variance and with the increase of block variance, the power of QTL detection decreased from 51.3% to 9.4%. In contrast to the RCBD method, the power of CIMLD and the adjusted mean method did not change for different block variances. The CIMLD method showed 1.2- to 7.6-fold higher power of QTL detection than the arithmetic or adjusted mean methods. Our proposed method was applied to real soybean (Glycine max) data as an example and 10 QTLs for biomass were identified that explained 65.87% of the phenotypic variation, while only three and two QTLs were identified by arithmetic and adjusted mean methods, respectively.
Ensemble codes involving hippocampal neurons are at risk during delayed performance tests.
Hampson, R E; Deadwyler, S A
1996-11-26
Multielectrode recording techniques were used to record ensemble activity from 10 to 16 simultaneously active CA1 and CA3 neurons in the rat hippocampus during performance of a spatial delayed-nonmatch-to-sample task. Extracted sources of variance were used to assess the nature of two different types of errors that accounted for 30% of total trials. The two types of errors included ensemble "miscodes" of sample phase information and errors associated with delay-dependent corruption or disappearance of sample information at the time of the nonmatch response. Statistical assessment of trial sequences and associated "strength" of hippocampal ensemble codes revealed that miscoded error trials always followed delay-dependent error trials in which encoding was "weak," indicating that the two types of errors were "linked." It was determined that the occurrence of weakly encoded, delay-dependent error trials initiated an ensemble encoding "strategy" that increased the chances of being correct on the next trial and avoided the occurrence of further delay-dependent errors. Unexpectedly, the strategy involved "strongly" encoding response position information from the prior (delay-dependent) error trial and carrying it forward to the sample phase of the next trial. This produced a miscode type error on trials in which the "carried over" information obliterated encoding of the sample phase response on the next trial. Application of this strategy, irrespective of outcome, was sufficient to reorient the animal to the proper between trial sequence of response contingencies (nonmatch-to-sample) and boost performance to 73% correct on subsequent trials. The capacity for ensemble analyses of strength of information encoding combined with statistical assessment of trial sequences therefore provided unique insight into the "dynamic" nature of the role hippocampus plays in delay type memory tasks.
Shi, Yun; Xu, Peiliang; Peng, Junhuan; Shi, Chuang; Liu, Jingnan
2014-01-01
Modern observation technology has verified that measurement errors can be proportional to the true values of measurements such as GPS, VLBI baselines and LiDAR. Observational models of this type are called multiplicative error models. This paper is to extend the work of Xu and Shimada published in 2000 on multiplicative error models to analytical error analysis of quantities of practical interest and estimates of the variance of unit weight. We analytically derive the variance-covariance matrices of the three least squares (LS) adjustments, the adjusted measurements and the corrections of measurements in multiplicative error models. For quality evaluation, we construct five estimators for the variance of unit weight in association of the three LS adjustment methods. Although LiDAR measurements are contaminated with multiplicative random errors, LiDAR-based digital elevation models (DEM) have been constructed as if they were of additive random errors. We will simulate a model landslide, which is assumed to be surveyed with LiDAR, and investigate the effect of LiDAR-type multiplicative error measurements on DEM construction and its effect on the estimate of landslide mass volume from the constructed DEM. PMID:24434880
NASA Technical Reports Server (NTRS)
Kimes, D. S.; Kerber, A. G.; Sellers, P. J.
1993-01-01
Spatial averaging errors which may occur when creating hemispherical reflectance maps for different cover types from direct nadir technique to estimate the hemispherical reflectance are assessed by comparing the results with those obtained with a knowledge-based system called VEG (Kimes et al., 1991, 1992). It was found that hemispherical reflectance errors provided by using VEG are much less than those using the direct nadir techniques, depending on conditions. Suggestions are made concerning sampling and averaging strategies for creating hemispherical reflectance maps for photosynthetic, carbon cycle, and climate change studies.
Comparison of structural and least-squares lines for estimating geologic relations
Williams, G.P.; Troutman, B.M.
1990-01-01
Two different goals in fitting straight lines to data are to estimate a "true" linear relation (physical law) and to predict values of the dependent variable with the smallest possible error. Regarding the first goal, a Monte Carlo study indicated that the structural-analysis (SA) method of fitting straight lines to data is superior to the ordinary least-squares (OLS) method for estimating "true" straight-line relations. Number of data points, slope and intercept of the true relation, and variances of the errors associated with the independent (X) and dependent (Y) variables influence the degree of agreement. For example, differences between the two line-fitting methods decrease as error in X becomes small relative to error in Y. Regarding the second goal-predicting the dependent variable-OLS is better than SA. Again, the difference diminishes as X takes on less error relative to Y. With respect to estimation of slope and intercept and prediction of Y, agreement between Monte Carlo results and large-sample theory was very good for sample sizes of 100, and fair to good for sample sizes of 20. The procedures and error measures are illustrated with two geologic examples. ?? 1990 International Association for Mathematical Geology.
NASA Astrophysics Data System (ADS)
Trung, Ha Duyen
2017-12-01
In this paper, the end-to-end performance of free-space optical (FSO) communication system combining with Amplify-and-Forward (AF)-assisted or fixed-gain relaying technology using subcarrier quadrature amplitude modulation (SC-QAM) over weak atmospheric turbulence channels modeled by log-normal distribution with pointing error impairments is studied. More specifically, unlike previous studies on AF relaying FSO communication systems without pointing error effects; the pointing error effect is studied by taking into account the influence of beamwidth, aperture size and jitter variance. In addition, a combination of these models to analyze the combined effect of atmospheric turbulence and pointing error to AF relaying FSO/SC-QAM systems is used. Finally, an analytical expression is derived to evaluate the average symbol error rate (ASER) performance of such systems. The numerical results show that the impact of pointing error on the performance of AF relaying FSO/SC-QAM systems and how we use proper values of aperture size and beamwidth to improve the performance of such systems. Some analytical results are confirmed by Monte-Carlo simulations.
NASA Technical Reports Server (NTRS)
Meneghini, Robert; Kim, Hyokyung
2016-01-01
For an airborne or spaceborne radar, the precipitation-induced path attenuation can be estimated from the measurements of the normalized surface cross section, sigma 0, in the presence and absence of precipitation. In one implementation, the mean rain-free estimate and its variability are found from a lookup table (LUT) derived from previously measured data. For the dual-frequency precipitation radar aboard the global precipitation measurement satellite, the nominal table consists of the statistics of the rain-free 0 over a 0.5 deg x 0.5 deg latitude-longitude grid using a three-month set of input data. However, a problem with the LUT is an insufficient number of samples in many cells. An alternative table is constructed by a stepwise procedure that begins with the statistics over a 0.25 deg x 0.25 deg grid. If the number of samples at a cell is too few, the area is expanded, cell by cell, choosing at each step that cell that minimizes the variance of the data. The question arises, however, as to whether the selected region corresponds to the smallest variance. To address this question, a second type of variable-averaging grid is constructed using all possible spatial configurations and computing the variance of the data within each region. Comparisons of the standard deviations for the fixed and variable-averaged grids are given as a function of incidence angle and surface type using a three-month set of data. The advantage of variable spatial averaging is that the average standard deviation can be reduced relative to the fixed grid while satisfying the minimum sample requirement.
Revised techniques for estimating peak discharges from channel width in Montana
Parrett, Charles; Hull, J.A.; Omang, R.J.
1987-01-01
This study was conducted to develop new estimating equations based on channel width and the updated flood frequency curves of previous investigations. Simple regression equations for estimating peak discharges with recurrence intervals of 2, 5, 10 , 25, 50, and 100 years were developed for seven regions in Montana. The standard errors of estimates for the equations that use active channel width as the independent variables ranged from 30% to 87%. The standard errors of estimate for the equations that use bankfull width as the independent variable ranged from 34% to 92%. The smallest standard errors generally occurred in the prediction equations for the 2-yr flood, 5-yr flood, and 10-yr flood, and the largest standard errors occurred in the prediction equations for the 100-yr flood. The equations that use active channel width and the equations that use bankfull width were determined to be about equally reliable in five regions. In the West Region, the equations that use bankfull width were slightly more reliable than those based on active channel width, whereas in the East-Central Region the equations that use active channel width were slightly more reliable than those based on bankfull width. Compared with similar equations previously developed, the standard errors of estimate for the new equations are substantially smaller in three regions and substantially larger in two regions. Limitations on the use of the estimating equations include: (1) The equations are based on stable conditions of channel geometry and prevailing water and sediment discharge; (2) The measurement of channel width requires a site visit, preferably by a person with experience in the method, and involves appreciable measurement errors; (3) Reliability of results from the equations for channel widths beyond the range of definition is unknown. In spite of the limitations, the estimating equations derived in this study are considered to be as reliable as estimating equations based on basin and climatic variables. Because the two types of estimating equations are independent, results from each can be weighted inversely proportional to their variances, and averaged. The weighted average estimate has a variance less than either individual estimate. (Author 's abstract)
Graf, Alexandra C; Bauer, Peter
2011-06-30
We calculate the maximum type 1 error rate of the pre-planned conventional fixed sample size test for comparing the means of independent normal distributions (with common known variance) which can be yielded when sample size and allocation rate to the treatment arms can be modified in an interim analysis. Thereby it is assumed that the experimenter fully exploits knowledge of the unblinded interim estimates of the treatment effects in order to maximize the conditional type 1 error rate. The 'worst-case' strategies require knowledge of the unknown common treatment effect under the null hypothesis. Although this is a rather hypothetical scenario it may be approached in practice when using a standard control treatment for which precise estimates are available from historical data. The maximum inflation of the type 1 error rate is substantially larger than derived by Proschan and Hunsberger (Biometrics 1995; 51:1315-1324) for design modifications applying balanced samples before and after the interim analysis. Corresponding upper limits for the maximum type 1 error rate are calculated for a number of situations arising from practical considerations (e.g. restricting the maximum sample size, not allowing sample size to decrease, allowing only increase in the sample size in the experimental treatment). The application is discussed for a motivating example. Copyright © 2011 John Wiley & Sons, Ltd.
Measurement System Characterization in the Presence of Measurement Errors
NASA Technical Reports Server (NTRS)
Commo, Sean A.
2012-01-01
In the calibration of a measurement system, data are collected in order to estimate a mathematical model between one or more factors of interest and a response. Ordinary least squares is a method employed to estimate the regression coefficients in the model. The method assumes that the factors are known without error; yet, it is implicitly known that the factors contain some uncertainty. In the literature, this uncertainty is known as measurement error. The measurement error affects both the estimates of the model coefficients and the prediction, or residual, errors. There are some methods, such as orthogonal least squares, that are employed in situations where measurement errors exist, but these methods do not directly incorporate the magnitude of the measurement errors. This research proposes a new method, known as modified least squares, that combines the principles of least squares with knowledge about the measurement errors. This knowledge is expressed in terms of the variance ratio - the ratio of response error variance to measurement error variance.
Wickenberg-Bolin, Ulrika; Göransson, Hanna; Fryknäs, Mårten; Gustafsson, Mats G; Isaksson, Anders
2006-03-13
Supervised learning for classification of cancer employs a set of design examples to learn how to discriminate between tumors. In practice it is crucial to confirm that the classifier is robust with good generalization performance to new examples, or at least that it performs better than random guessing. A suggested alternative is to obtain a confidence interval of the error rate using repeated design and test sets selected from available examples. However, it is known that even in the ideal situation of repeated designs and tests with completely novel samples in each cycle, a small test set size leads to a large bias in the estimate of the true variance between design sets. Therefore different methods for small sample performance estimation such as a recently proposed procedure called Repeated Random Sampling (RSS) is also expected to result in heavily biased estimates, which in turn translates into biased confidence intervals. Here we explore such biases and develop a refined algorithm called Repeated Independent Design and Test (RIDT). Our simulations reveal that repeated designs and tests based on resampling in a fixed bag of samples yield a biased variance estimate. We also demonstrate that it is possible to obtain an improved variance estimate by means of a procedure that explicitly models how this bias depends on the number of samples used for testing. For the special case of repeated designs and tests using new samples for each design and test, we present an exact analytical expression for how the expected value of the bias decreases with the size of the test set. We show that via modeling and subsequent reduction of the small sample bias, it is possible to obtain an improved estimate of the variance of classifier performance between design sets. However, the uncertainty of the variance estimate is large in the simulations performed indicating that the method in its present form cannot be directly applied to small data sets.
Aulenbach, Brent T.; Burns, Douglas A.; Shanley, James B.; Yanai, Ruth D.; Bae, Kikang; Wild, Adam; Yang, Yang; Yi, Dong
2016-01-01
Estimating streamwater solute loads is a central objective of many water-quality monitoring and research studies, as loads are used to compare with atmospheric inputs, to infer biogeochemical processes, and to assess whether water quality is improving or degrading. In this study, we evaluate loads and associated errors to determine the best load estimation technique among three methods (a period-weighted approach, the regression-model method, and the composite method) based on a solute's concentration dynamics and sampling frequency. We evaluated a broad range of varying concentration dynamics with stream flow and season using four dissolved solutes (sulfate, silica, nitrate, and dissolved organic carbon) at five diverse small watersheds (Sleepers River Research Watershed, VT; Hubbard Brook Experimental Forest, NH; Biscuit Brook Watershed, NY; Panola Mountain Research Watershed, GA; and Río Mameyes Watershed, PR) with fairly high-frequency sampling during a 10- to 11-yr period. Data sets with three different sampling frequencies were derived from the full data set at each site (weekly plus storm/snowmelt events, weekly, and monthly) and errors in loads were assessed for the study period, annually, and monthly. For solutes that had a moderate to strong concentration–discharge relation, the composite method performed best, unless the autocorrelation of the model residuals was <0.2, in which case the regression-model method was most appropriate. For solutes that had a nonexistent or weak concentration–discharge relation (modelR2 < about 0.3), the period-weighted approach was most appropriate. The lowest errors in loads were achieved for solutes with the strongest concentration–discharge relations. Sample and regression model diagnostics could be used to approximate overall accuracies and annual precisions. For the period-weighed approach, errors were lower when the variance in concentrations was lower, the degree of autocorrelation in the concentrations was higher, and sampling frequency was higher. The period-weighted approach was most sensitive to sampling frequency. For the regression-model and composite methods, errors were lower when the variance in model residuals was lower. For the composite method, errors were lower when the autocorrelation in the residuals was higher. Guidelines to determine the best load estimation method based on solute concentration–discharge dynamics and diagnostics are presented, and should be applicable to other studies.
Biostatistics Series Module 5: Determining Sample Size
Hazra, Avijit; Gogtay, Nithya
2016-01-01
Determining the appropriate sample size for a study, whatever be its type, is a fundamental aspect of biomedical research. An adequate sample ensures that the study will yield reliable information, regardless of whether the data ultimately suggests a clinically important difference between the interventions or elements being studied. The probability of Type 1 and Type 2 errors, the expected variance in the sample and the effect size are the essential determinants of sample size in interventional studies. Any method for deriving a conclusion from experimental data carries with it some risk of drawing a false conclusion. Two types of false conclusion may occur, called Type 1 and Type 2 errors, whose probabilities are denoted by the symbols σ and β. A Type 1 error occurs when one concludes that a difference exists between the groups being compared when, in reality, it does not. This is akin to a false positive result. A Type 2 error occurs when one concludes that difference does not exist when, in reality, a difference does exist, and it is equal to or larger than the effect size defined by the alternative to the null hypothesis. This may be viewed as a false negative result. When considering the risk of Type 2 error, it is more intuitive to think in terms of power of the study or (1 − β). Power denotes the probability of detecting a difference when a difference does exist between the groups being compared. Smaller α or larger power will increase sample size. Conventional acceptable values for power and α are 80% or above and 5% or below, respectively, when calculating sample size. Increasing variance in the sample tends to increase the sample size required to achieve a given power level. The effect size is the smallest clinically important difference that is sought to be detected and, rather than statistical convention, is a matter of past experience and clinical judgment. Larger samples are required if smaller differences are to be detected. Although the principles are long known, historically, sample size determination has been difficult, because of relatively complex mathematical considerations and numerous different formulas. However, of late, there has been remarkable improvement in the availability, capability, and user-friendliness of power and sample size determination software. Many can execute routines for determination of sample size and power for a wide variety of research designs and statistical tests. With the drudgery of mathematical calculation gone, researchers must now concentrate on determining appropriate sample size and achieving these targets, so that study conclusions can be accepted as meaningful. PMID:27688437
Evaluation of three lidar scanning strategies for turbulence measurements
Newman, Jennifer F.; Klein, Petra M.; Wharton, Sonia; ...
2016-05-03
Several errors occur when a traditional Doppler beam swinging (DBS) or velocity–azimuth display (VAD) strategy is used to measure turbulence with a lidar. To mitigate some of these errors, a scanning strategy was recently developed which employs six beam positions to independently estimate the u, v, and w velocity variances and covariances. In order to assess the ability of these different scanning techniques to measure turbulence, a Halo scanning lidar, WindCube v2 pulsed lidar, and ZephIR continuous wave lidar were deployed at field sites in Oklahoma and Colorado with collocated sonic anemometers.Results indicate that the six-beam strategy mitigates some of the errors caused bymore » VAD and DBS scans, but the strategy is strongly affected by errors in the variance measured at the different beam positions. The ZephIR and WindCube lidars overestimated horizontal variance values by over 60 % under unstable conditions as a result of variance contamination, where additional variance components contaminate the true value of the variance. A correction method was developed for the WindCube lidar that uses variance calculated from the vertical beam position to reduce variance contamination in the u and v variance components. The correction method reduced WindCube variance estimates by over 20 % at both the Oklahoma and Colorado sites under unstable conditions, when variance contamination is largest. This correction method can be easily applied to other lidars that contain a vertical beam position and is a promising method for accurately estimating turbulence with commercially available lidars.« less
Evaluation of three lidar scanning strategies for turbulence measurements
DOE Office of Scientific and Technical Information (OSTI.GOV)
Newman, Jennifer F.; Klein, Petra M.; Wharton, Sonia
Several errors occur when a traditional Doppler beam swinging (DBS) or velocity–azimuth display (VAD) strategy is used to measure turbulence with a lidar. To mitigate some of these errors, a scanning strategy was recently developed which employs six beam positions to independently estimate the u, v, and w velocity variances and covariances. In order to assess the ability of these different scanning techniques to measure turbulence, a Halo scanning lidar, WindCube v2 pulsed lidar, and ZephIR continuous wave lidar were deployed at field sites in Oklahoma and Colorado with collocated sonic anemometers.Results indicate that the six-beam strategy mitigates some of the errors caused bymore » VAD and DBS scans, but the strategy is strongly affected by errors in the variance measured at the different beam positions. The ZephIR and WindCube lidars overestimated horizontal variance values by over 60 % under unstable conditions as a result of variance contamination, where additional variance components contaminate the true value of the variance. A correction method was developed for the WindCube lidar that uses variance calculated from the vertical beam position to reduce variance contamination in the u and v variance components. The correction method reduced WindCube variance estimates by over 20 % at both the Oklahoma and Colorado sites under unstable conditions, when variance contamination is largest. This correction method can be easily applied to other lidars that contain a vertical beam position and is a promising method for accurately estimating turbulence with commercially available lidars.« less
Comparisons of refractive errors between twins and singletons in Chinese school-age samples.
Hur, Yoon-Mi; Zheng, Yingfeng; Huang, Wenyong; Ding, Xiaohu; He, Mingguang
2009-02-01
Studies have reported that refractive errors are associated with premature births. As twins have higher prevalence of prematurity than singletons, it is important to assess similarity of the prevalence of refractive errors in twins and singletons for proper interpretations and generalizations of the findings from twin studies. We compared refractive errors and diopter hours between 561 pairs of twins and 3757 singletons who are representative of school-age children (7-15 years) residing in an urban area of southern China. We found that the means and variances of the continuous measurement of spherical equivalent refractive error and diopter hours were not significantly different between twins and singletons. Although the prevalence of myopia was comparable between twins and singletons, that of hyperopia and astigmatism was slightly but significantly higher in twins than in singletons. These results are inconsistent with those of adult studies that showed no differences in refractive errors between twins and singletons. Given that the sample size of twins is relatively small and that this study is the first to demonstrate minor differences in refractive errors between twins and singletons, future replications are necessary to determine whether the slightly higher prevalence of refractive errors in twins than in singletons found in this study was due to a sampling error or to the developmental delay often observed in twins in childhood.
Data assimilation method based on the constraints of confidence region
NASA Astrophysics Data System (ADS)
Li, Yong; Li, Siming; Sheng, Yao; Wang, Luheng
2018-03-01
The ensemble Kalman filter (EnKF) is a distinguished data assimilation method that is widely used and studied in various fields including methodology and oceanography. However, due to the limited sample size or imprecise dynamics model, it is usually easy for the forecast error variance to be underestimated, which further leads to the phenomenon of filter divergence. Additionally, the assimilation results of the initial stage are poor if the initial condition settings differ greatly from the true initial state. To address these problems, the variance inflation procedure is usually adopted. In this paper, we propose a new method based on the constraints of a confidence region constructed by the observations, called EnCR, to estimate the inflation parameter of the forecast error variance of the EnKF method. In the new method, the state estimate is more robust to both the inaccurate forecast models and initial condition settings. The new method is compared with other adaptive data assimilation methods in the Lorenz-63 and Lorenz-96 models under various model parameter settings. The simulation results show that the new method performs better than the competing methods.
Integrating mean and variance heterogeneities to identify differentially expressed genes.
Ouyang, Weiwei; An, Qiang; Zhao, Jinying; Qin, Huaizhen
2016-12-06
In functional genomics studies, tests on mean heterogeneity have been widely employed to identify differentially expressed genes with distinct mean expression levels under different experimental conditions. Variance heterogeneity (aka, the difference between condition-specific variances) of gene expression levels is simply neglected or calibrated for as an impediment. The mean heterogeneity in the expression level of a gene reflects one aspect of its distribution alteration; and variance heterogeneity induced by condition change may reflect another aspect. Change in condition may alter both mean and some higher-order characteristics of the distributions of expression levels of susceptible genes. In this report, we put forth a conception of mean-variance differentially expressed (MVDE) genes, whose expression means and variances are sensitive to the change in experimental condition. We mathematically proved the null independence of existent mean heterogeneity tests and variance heterogeneity tests. Based on the independence, we proposed an integrative mean-variance test (IMVT) to combine gene-wise mean heterogeneity and variance heterogeneity induced by condition change. The IMVT outperformed its competitors under comprehensive simulations of normality and Laplace settings. For moderate samples, the IMVT well controlled type I error rates, and so did existent mean heterogeneity test (i.e., the Welch t test (WT), the moderated Welch t test (MWT)) and the procedure of separate tests on mean and variance heterogeneities (SMVT), but the likelihood ratio test (LRT) severely inflated type I error rates. In presence of variance heterogeneity, the IMVT appeared noticeably more powerful than all the valid mean heterogeneity tests. Application to the gene profiles of peripheral circulating B raised solid evidence of informative variance heterogeneity. After adjusting for background data structure, the IMVT replicated previous discoveries and identified novel experiment-wide significant MVDE genes. Our results indicate tremendous potential gain of integrating informative variance heterogeneity after adjusting for global confounders and background data structure. The proposed informative integration test better summarizes the impacts of condition change on expression distributions of susceptible genes than do the existent competitors. Therefore, particular attention should be paid to explicitly exploit the variance heterogeneity induced by condition change in functional genomics analysis.
ERIC Educational Resources Information Center
Bauer, Daniel J.; Preacher, Kristopher J.; Gil, Karen M.
2006-01-01
The authors propose new procedures for evaluating direct, indirect, and total effects in multilevel models when all relevant variables are measured at Level 1 and all effects are random. Formulas are provided for the mean and variance of the indirect and total effects and for the sampling variances of the average indirect and total effects.…
Mapping from disease-specific measures to health-state utility values in individuals with migraine.
Gillard, Patrick J; Devine, Beth; Varon, Sepideh F; Liu, Lei; Sullivan, Sean D
2012-05-01
The objective of this study was to develop empirical algorithms that estimate health-state utility values from disease-specific quality-of-life scores in individuals with migraine. Data from a cross-sectional, multicountry study were used. Individuals with episodic and chronic migraine were randomly assigned to training or validation samples. Spearman's correlation coefficients between paired EuroQol five-dimensional (EQ-5D) questionnaire utility values and both Headache Impact Test (HIT-6) scores and Migraine-Specific Quality-of-Life Questionnaire version 2.1 (MSQ) domain scores (role restrictive, role preventive, and emotional function) were examined. Regression models were constructed to estimate EQ-5D questionnaire utility values from the HIT-6 score or the MSQ domain scores. Preferred algorithms were confirmed in the validation samples. In episodic migraine, the preferred HIT-6 and MSQ algorithms explained 22% and 25% of the variance (R(2)) in the training samples, respectively, and had similar prediction errors (root mean square errors of 0.30). In chronic migraine, the preferred HIT-6 and MSQ algorithms explained 36% and 45% of the variance in the training samples, respectively, and had similar prediction errors (root mean square errors 0.31 and 0.29). In episodic and chronic migraine, no statistically significant differences were observed between the mean observed and the mean estimated EQ-5D questionnaire utility values for the preferred HIT-6 and MSQ algorithms in the validation samples. The relationship between the EQ-5D questionnaire and the HIT-6 or the MSQ is adequate to use regression equations to estimate EQ-5D questionnaire utility values. The preferred HIT-6 and MSQ algorithms will be useful in estimating health-state utilities in migraine trials in which no preference-based measure is present. Copyright © 2012 International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. All rights reserved.
NASA Astrophysics Data System (ADS)
Caimmi, R.
2011-08-01
Concerning bivariate least squares linear regression, the classical approach pursued for functional models in earlier attempts ( York, 1966, 1969) is reviewed using a new formalism in terms of deviation (matrix) traces which, for unweighted data, reduce to usual quantities leaving aside an unessential (but dimensional) multiplicative factor. Within the framework of classical error models, the dependent variable relates to the independent variable according to the usual additive model. The classes of linear models considered are regression lines in the general case of correlated errors in X and in Y for weighted data, and in the opposite limiting situations of (i) uncorrelated errors in X and in Y, and (ii) completely correlated errors in X and in Y. The special case of (C) generalized orthogonal regression is considered in detail together with well known subcases, namely: (Y) errors in X negligible (ideally null) with respect to errors in Y; (X) errors in Y negligible (ideally null) with respect to errors in X; (O) genuine orthogonal regression; (R) reduced major-axis regression. In the limit of unweighted data, the results determined for functional models are compared with their counterparts related to extreme structural models i.e. the instrumental scatter is negligible (ideally null) with respect to the intrinsic scatter ( Isobe et al., 1990; Feigelson and Babu, 1992). While regression line slope and intercept estimators for functional and structural models necessarily coincide, the contrary holds for related variance estimators even if the residuals obey a Gaussian distribution, with the exception of Y models. An example of astronomical application is considered, concerning the [O/H]-[Fe/H] empirical relations deduced from five samples related to different stars and/or different methods of oxygen abundance determination. For selected samples and assigned methods, different regression models yield consistent results within the errors (∓ σ) for both heteroscedastic and homoscedastic data. Conversely, samples related to different methods produce discrepant results, due to the presence of (still undetected) systematic errors, which implies no definitive statement can be made at present. A comparison is also made between different expressions of regression line slope and intercept variance estimators, where fractional discrepancies are found to be not exceeding a few percent, which grows up to about 20% in the presence of large dispersion data. An extension of the formalism to structural models is left to a forthcoming paper.
Phylogenomics of Lophotrochozoa with Consideration of Systematic Error.
Kocot, Kevin M; Struck, Torsten H; Merkel, Julia; Waits, Damien S; Todt, Christiane; Brannock, Pamela M; Weese, David A; Cannon, Johanna T; Moroz, Leonid L; Lieb, Bernhard; Halanych, Kenneth M
2017-03-01
Phylogenomic studies have improved understanding of deep metazoan phylogeny and show promise for resolving incongruences among analyses based on limited numbers of loci. One region of the animal tree that has been especially difficult to resolve, even with phylogenomic approaches, is relationships within Lophotrochozoa (the animal clade that includes molluscs, annelids, and flatworms among others). Lack of resolution in phylogenomic analyses could be due to insufficient phylogenetic signal, limitations in taxon and/or gene sampling, or systematic error. Here, we investigated why lophotrochozoan phylogeny has been such a difficult question to answer by identifying and reducing sources of systematic error. We supplemented existing data with 32 new transcriptomes spanning the diversity of Lophotrochozoa and constructed a new set of Lophotrochozoa-specific core orthologs. Of these, 638 orthologous groups (OGs) passed strict screening for paralogy using a tree-based approach. In order to reduce possible sources of systematic error, we calculated branch-length heterogeneity, evolutionary rate, percent missing data, compositional bias, and saturation for each OG and analyzed increasingly stricter subsets of only the most stringent (best) OGs for these five variables. Principal component analysis of the values for each factor examined for each OG revealed that compositional heterogeneity and average patristic distance contributed most to the variance observed along the first principal component while branch-length heterogeneity and, to a lesser extent, saturation contributed most to the variance observed along the second. Missing data did not strongly contribute to either. Additional sensitivity analyses examined effects of removing taxa with heterogeneous branch lengths, large amounts of missing data, and compositional heterogeneity. Although our analyses do not unambiguously resolve lophotrochozoan phylogeny, we advance the field by reducing the list of viable hypotheses. Moreover, our systematic approach for dissection of phylogenomic data can be applied to explore sources of incongruence and poor support in any phylogenomic data set. [Annelida; Brachiopoda; Bryozoa; Entoprocta; Mollusca; Nemertea; Phoronida; Platyzoa; Polyzoa; Spiralia; Trochozoa.]. © The Author(s) 2016. Published by Oxford University Press, on behalf of the Society of Systematic Biologists. All rights reserved. For Permissions, please email: journals.permissions@oup.com.
Representativeness of laboratory sampling procedures for the analysis of trace metals in soil.
Dubé, Jean-Sébastien; Boudreault, Jean-Philippe; Bost, Régis; Sona, Mirela; Duhaime, François; Éthier, Yannic
2015-08-01
This study was conducted to assess the representativeness of laboratory sampling protocols for purposes of trace metal analysis in soil. Five laboratory protocols were compared, including conventional grab sampling, to assess the influence of sectorial splitting, sieving, and grinding on measured trace metal concentrations and their variability. It was concluded that grinding was the most important factor in controlling the variability of trace metal concentrations. Grinding increased the reproducibility of sample mass reduction by rotary sectorial splitting by up to two orders of magnitude. Combined with rotary sectorial splitting, grinding increased the reproducibility of trace metal concentrations by almost three orders of magnitude compared to grab sampling. Moreover, results showed that if grinding is used as part of a mass reduction protocol by sectorial splitting, the effect of sieving on reproducibility became insignificant. Gy's sampling theory and practice was also used to analyze the aforementioned sampling protocols. While the theoretical relative variances calculated for each sampling protocol qualitatively agreed with the experimental variances, their quantitative agreement was very poor. It was assumed that the parameters used in the calculation of theoretical sampling variances may not correctly estimate the constitutional heterogeneity of soils or soil-like materials. Finally, the results have highlighted the pitfalls of grab sampling, namely, the fact that it does not exert control over incorrect sampling errors and that it is strongly affected by distribution heterogeneity.
Field and laboratory procedures used in a soil chronosequence study
Singer, Michael J.; Janitzky, Peter
1986-01-01
In 1978, the late Denis Marchand initiated a research project entitled "Soil Correlation and Dating at the U.S. Geological Survey" to determine the usefulness of soils in solving geologic problems. Marchand proposed to establish soil chronosequences that could be dated independently of soil development by using radiometric and other numeric dating methods. In addition, by comparing dated chronosequences in different environments, rates of soil development could be studied and compared among varying climates and mineralogical conditions. The project was fundamental in documenting the value of soils in studies of mapping, correlating, and dating late Cenozoic deposits and in studying soil genesis. All published reports by members of the project are included in the bibliography.The project demanded that methods be adapted or developed to ensure comparability over a wide variation in soil types. Emphasis was placed on obtaining professional expertise and on establishing consistent techniques, especially for the field, laboratory, and data-compilation methods. Since 1978, twelve chronosequences have been sampled and analyzed by members of this project, and methods have been established and used consistently for analysis of the samples.The goals of this report are to:Document the methods used for the study on soil chronosequences,Present the results of tests that were run for precision, accuracy, and effectiveness, andDiscuss our modifications to standard procedures.Many of the methods presented herein are standard and have been reported elsewhere. However, we assume less prior analytical knowledge in our descriptions; thus, the manual should be easy to follow for the inexperienced analyst. Each chapter presents one or more references of the basic principle, an equipment and reagents list, and the detailed procedure. In some chapters this is followed by additional remarks or example calculations.The flow diagram in figure 1 outlines the step-by-step procedures used to obtain and analyze soil samples for this study. The soils analyzed had a wide range of characteristics (such as clay content, mineralogy, salinity, and acidity). Initially, a major task was to test and select methods that could be applied and interpreted similarly for the various types of soils. Tests were conducted to establish the effectiveness and comparability of analytical techniques, and the data for such tests are included in figures, tables, and discussions. In addition, many replicate analyses of samples have established a "standard error" or "coefficient of variance" which indicates the average reproducibility of each laboratory procedure. These averaged errors are reported as percentage of a given value. For example, in particle-size determination, 3 percent error for 10 percent clay content equals 10 ± 0.3 percent clay. The error sources were examined to determine, for example, if the error in particle-size determination was dependent on clay content. No such biases were found, and data are reported as percent error in the text and in tables of reproducibility.
An approach to the analysis of performance of quasi-optimum digital phase-locked loops.
NASA Technical Reports Server (NTRS)
Polk, D. R.; Gupta, S. C.
1973-01-01
An approach to the analysis of performance of quasi-optimum digital phase-locked loops (DPLL's) is presented. An expression for the characteristic function of the prior error in the state estimate is derived, and from this expression an infinite dimensional equation for the prior error variance is obtained. The prior error-variance equation is a function of the communication system model and the DPLL gain and is independent of the method used to derive the DPLL gain. Two approximations are discussed for reducing the prior error-variance equation to finite dimension. The effectiveness of one approximation in analyzing DPLL performance is studied.
Radiometric errors in complex Fourier transform spectrometry.
Sromovsky, Lawrence A
2003-04-01
A complex spectrum arises from the Fourier transform of an asymmetric interferogram. A rigorous derivation shows that the rms noise in the real part of that spectrum is indeed given by the commonly used relation sigmaR = 2X x NEP/(etaAomega square root(tauN)), where NEP is the delay-independent and uncorrelated detector noise-equivalent power per unit bandwidth, +/- X is the delay range measured with N samples averaging for a time tau per sample, eta is the system optical efficiency, and Aomega is the system throughput. A real spectrum produced by complex calibration with two complex reference spectra [Appl. Opt. 27, 3210 (1988)] has a variance sigmaL2 = sigmaR2 + sigma(c)2 (Lh - Ls)2/(Lh - Lc)2 + sigma(h)2 (Ls - Lc)2/(Lh - Lc)2, valid for sigmaR, sigma(c), and sigma(h) small compared with Lh - Lc, where Ls, Lh, and Lc are scene, hot reference, and cold reference spectra, respectively, and sigma(c) and sigma(h) are the respective combined uncertainties in knowledge and measurement of the hot and cold reference spectra.
On the Power of Multivariate Latent Growth Curve Models to Detect Correlated Change
ERIC Educational Resources Information Center
Hertzog, Christopher; Lindenberger, Ulman; Ghisletta, Paolo; Oertzen, Timo von
2006-01-01
We evaluated the statistical power of single-indicator latent growth curve models (LGCMs) to detect correlated change between two variables (covariance of slopes) as a function of sample size, number of longitudinal measurement occasions, and reliability (measurement error variance). Power approximations following the method of Satorra and Saris…
NASA Technical Reports Server (NTRS)
Kundu, Prasun K.; Bell, T. L.; Lau, William K. M. (Technical Monitor)
2002-01-01
A characteristic feature of rainfall statistics is that they in general depend on the space and time scales over which rain data are averaged. As a part of an earlier effort to determine the sampling error of satellite rain averages, a space-time model of rainfall statistics was developed to describe the statistics of gridded rain observed in GATE. The model allows one to compute the second moment statistics of space- and time-averaged rain rate which can be fitted to satellite or rain gauge data to determine the four model parameters appearing in the precipitation spectrum - an overall strength parameter, a characteristic length separating the long and short wavelength regimes and a characteristic relaxation time for decay of the autocorrelation of the instantaneous local rain rate and a certain 'fractal' power law exponent. For area-averaged instantaneous rain rate, this exponent governs the power law dependence of these statistics on the averaging length scale $L$ predicted by the model in the limit of small $L$. In particular, the variance of rain rate averaged over an $L \\times L$ area exhibits a power law singularity as $L \\rightarrow 0$. In the present work the model is used to investigate how the statistics of area-averaged rain rate over the tropical Western Pacific measured with ship borne radar during TOGA COARE (Tropical Ocean Global Atmosphere Coupled Ocean Atmospheric Response Experiment) and gridded on a 2 km grid depends on the size of the spatial averaging scale. Good agreement is found between the data and predictions from the model over a wide range of averaging length scales.
Counting OCR errors in typeset text
NASA Astrophysics Data System (ADS)
Sandberg, Jonathan S.
1995-03-01
Frequently object recognition accuracy is a key component in the performance analysis of pattern matching systems. In the past three years, the results of numerous excellent and rigorous studies of OCR system typeset-character accuracy (henceforth OCR accuracy) have been published, encouraging performance comparisons between a variety of OCR products and technologies. These published figures are important; OCR vendor advertisements in the popular trade magazines lead readers to believe that published OCR accuracy figures effect market share in the lucrative OCR market. Curiously, a detailed review of many of these OCR error occurrence counting results reveals that they are not reproducible as published and they are not strictly comparable due to larger variances in the counts than would be expected by the sampling variance. Naturally, since OCR accuracy is based on a ratio of the number of OCR errors over the size of the text searched for errors, imprecise OCR error accounting leads to similar imprecision in OCR accuracy. Some published papers use informal, non-automatic, or intuitively correct OCR error accounting. Still other published results present OCR error accounting methods based on string matching algorithms such as dynamic programming using Levenshtein (edit) distance but omit critical implementation details (such as the existence of suspect markers in the OCR generated output or the weights used in the dynamic programming minimization procedure). The problem with not specifically revealing the accounting method is that the number of errors found by different methods are significantly different. This paper identifies the basic accounting methods used to measure OCR errors in typeset text and offers an evaluation and comparison of the various accounting methods.
Klompmaker, Jochem O; Montagne, Denise R; Meliefste, Kees; Hoek, Gerard; Brunekreef, Bert
2015-03-01
Recently, short-term monitoring campaigns have been carried out to investigate the spatial variation of air pollutants within cities. Typically, such campaigns are based on short-term measurements at relatively large numbers of locations. It is largely unknown how well these studies capture the spatial variation of long term average concentrations. The aim of this study was to evaluate the within-site temporal and between-site spatial variation of the concentration of ultrafine particles (UFPs) and black carbon (BC) in a short-term monitoring campaign. In Amsterdam and Rotterdam (the Netherlands) measurements of number counts of particles larger than 10nm as a surrogate for UFP and BC were performed at 80 sites per city. Each site was measured in three different seasons of 2013 (winter, spring, summer). Sites were selected from busy urban streets, urban background, regional background and near highways, waterways and green areas, to obtain sufficient spatial contrast. Continuous measurements were performed for 30 min per site between 9 and 16 h to avoid traffic spikes of the rush hour. Concentrations were simultaneously measured at a reference site to correct for temporal variation. We calculated within- and between-site variance components reflecting temporal and spatial variations. Variance ratios were compared with previous campaigns with longer sampling durations per sample (24h to 14 days). The within-site variance was 2.17 and 2.44 times higher than the between-site variance for UFP and BC, respectively. In two previous studies based upon longer sampling duration much smaller variance ratios were found (0.31 and 0.09 for UFP and BC). Correction for temporal variation from a reference site was less effective for the short-term monitoring campaign compared to the campaigns with longer duration. Concentrations of BC and UFP were on average 1.6 and 1.5 times higher at urban street compared to urban background sites. No significant differences between the other site types and urban background were found. The high within to between-site concentration variances may result in the loss of precision and low explained variance when average concentrations from short-term campaigns are used to develop land use regression models. Copyright © 2014 Elsevier B.V. All rights reserved.
Geostatistical modeling of riparian forest microclimate and its implications for sampling
Eskelson, B.N.I.; Anderson, P.D.; Hagar, J.C.; Temesgen, H.
2011-01-01
Predictive models of microclimate under various site conditions in forested headwater stream - riparian areas are poorly developed, and sampling designs for characterizing underlying riparian microclimate gradients are sparse. We used riparian microclimate data collected at eight headwater streams in the Oregon Coast Range to compare ordinary kriging (OK), universal kriging (UK), and kriging with external drift (KED) for point prediction of mean maximum air temperature (Tair). Several topographic and forest structure characteristics were considered as site-specific parameters. Height above stream and distance to stream were the most important covariates in the KED models, which outperformed OK and UK in terms of root mean square error. Sample patterns were optimized based on the kriging variance and the weighted means of shortest distance criterion using the simulated annealing algorithm. The optimized sample patterns outperformed systematic sample patterns in terms of mean kriging variance mainly for small sample sizes. These findings suggest methods for increasing efficiency of microclimate monitoring in riparian areas.
Evaluation of statistical models for forecast errors from the HBV model
NASA Astrophysics Data System (ADS)
Engeland, Kolbjørn; Renard, Benjamin; Steinsland, Ingelin; Kolberg, Sjur
2010-04-01
SummaryThree statistical models for the forecast errors for inflow into the Langvatn reservoir in Northern Norway have been constructed and tested according to the agreement between (i) the forecast distribution and the observations and (ii) median values of the forecast distribution and the observations. For the first model observed and forecasted inflows were transformed by the Box-Cox transformation before a first order auto-regressive model was constructed for the forecast errors. The parameters were conditioned on weather classes. In the second model the Normal Quantile Transformation (NQT) was applied on observed and forecasted inflows before a similar first order auto-regressive model was constructed for the forecast errors. For the third model positive and negative errors were modeled separately. The errors were first NQT-transformed before conditioning the mean error values on climate, forecasted inflow and yesterday's error. To test the three models we applied three criterions: we wanted (a) the forecast distribution to be reliable; (b) the forecast intervals to be narrow; (c) the median values of the forecast distribution to be close to the observed values. Models 1 and 2 gave almost identical results. The median values improved the forecast with Nash-Sutcliffe R eff increasing from 0.77 for the original forecast to 0.87 for the corrected forecasts. Models 1 and 2 over-estimated the forecast intervals but gave the narrowest intervals. Their main drawback was that the distributions are less reliable than Model 3. For Model 3 the median values did not fit well since the auto-correlation was not accounted for. Since Model 3 did not benefit from the potential variance reduction that lies in bias estimation and removal it gave on average wider forecasts intervals than the two other models. At the same time Model 3 on average slightly under-estimated the forecast intervals, probably explained by the use of average measures to evaluate the fit.
Observed spatiotemporal variability of boundary-layer turbulence over flat, heterogeneous terrain
NASA Astrophysics Data System (ADS)
Maurer, V.; Kalthoff, N.; Wieser, A.; Kohler, M.; Mauder, M.; Gantner, L.
2016-02-01
In the spring of 2013, extensive measurements with multiple Doppler lidar systems were performed. The instruments were arranged in a triangle with edge lengths of about 3 km in a moderately flat, agriculturally used terrain in northwestern Germany. For 6 mostly cloud-free convective days, vertical velocity variance profiles were calculated. Weighted-averaged surface fluxes proved to be more appropriate than data from individual sites for scaling the variance profiles; but even then, the scatter of profiles was mostly larger than the statistical error. The scatter could not be explained by mean wind speed or stability, whereas time periods with significantly increased variance contained broader thermals. Periods with an elevated maximum of the variance profiles could also be related to broad thermals. Moreover, statistically significant spatial differences of variance were found. They were not influenced by the existing surface heterogeneity. Instead, thermals were preserved between two sites when the travel time was shorter than the large-eddy turnover time. At the same time, no thermals passed for more than 2 h at a third site that was located perpendicular to the mean wind direction in relation to the first two sites. Organized structures of turbulence with subsidence prevailing in the surroundings of thermals can thus partly explain significant spatial variance differences existing for several hours. Therefore, the representativeness of individual variance profiles derived from measurements at a single site cannot be assumed.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Beer, M.
1980-12-01
The maximum likelihood method for the multivariate normal distribution is applied to the case of several individual eigenvalues. Correlated Monte Carlo estimates of the eigenvalue are assumed to follow this prescription and aspects of the assumption are examined. Monte Carlo cell calculations using the SAM-CE and VIM codes for the TRX-1 and TRX-2 benchmark reactors, and SAM-CE full core results are analyzed with this method. Variance reductions of a few percent to a factor of 2 are obtained from maximum likelihood estimation as compared with the simple average and the minimum variance individual eigenvalue. The numerical results verify that themore » use of sample variances and correlation coefficients in place of the corresponding population statistics still leads to nearly minimum variance estimation for a sufficient number of histories and aggregates.« less
Wang, Ping; Liu, Xiaoxia; Cao, Tian; Fu, Huihua; Wang, Ranran; Guo, Lixin
2016-09-20
The impact of nonzero boresight pointing errors on the system performance of decode-and-forward protocol-based multihop parallel optical wireless communication systems is studied. For the aggregated fading channel, the atmospheric turbulence is simulated by an exponentiated Weibull model, and pointing errors are described by one recently proposed statistical model including both boresight and jitter. The binary phase-shift keying subcarrier intensity modulation-based analytical average bit error rate (ABER) and outage probability expressions are achieved for a nonidentically and independently distributed system. The ABER and outage probability are then analyzed with different turbulence strengths, receiving aperture sizes, structure parameters (P and Q), jitter variances, and boresight displacements. The results show that aperture averaging offers almost the same system performance improvement with boresight included or not, despite the values of P and Q. The performance enhancement owing to the increase of cooperative path (P) is more evident with nonzero boresight than that with zero boresight (jitter only), whereas the performance deterioration because of the increasing hops (Q) with nonzero boresight is almost the same as that with zero boresight. Monte Carlo simulation is offered to verify the validity of ABER and outage probability expressions.
Decision Support System for ASD (Aeronatical Systems Division) Program Managers.
1985-09-01
and TA. Lastly, but definitely not least, I wish to thank my wife Stephanie and our three girls . During this arduous undertaking, many times...necessary to my work. Our girls carried both of us thru the AFIT experience. The cheerful and playful pursuits of Shaye and Jessica pierced the doldrums of...an average of .93. This high coefficient means very little of the variance found in the results of their survey is due to measurement error (8:536
Hedging Your Bets by Learning Reward Correlations in the Human Brain
Wunderlich, Klaus; Symmonds, Mkael; Bossaerts, Peter; Dolan, Raymond J.
2011-01-01
Summary Human subjects are proficient at tracking the mean and variance of rewards and updating these via prediction errors. Here, we addressed whether humans can also learn about higher-order relationships between distinct environmental outcomes, a defining ecological feature of contexts where multiple sources of rewards are available. By manipulating the degree to which distinct outcomes are correlated, we show that subjects implemented an explicit model-based strategy to learn the associated outcome correlations and were adept in using that information to dynamically adjust their choices in a task that required a minimization of outcome variance. Importantly, the experimentally generated outcome correlations were explicitly represented neuronally in right midinsula with a learning prediction error signal expressed in rostral anterior cingulate cortex. Thus, our data show that the human brain represents higher-order correlation structures between rewards, a core adaptive ability whose immediate benefit is optimized sampling. PMID:21943609
Sampling Variances and Covariances of Parameter Estimates in Item Response Theory.
1982-08-01
substituting (15) into (16) and solving for k and K k = b b1 - o K , (17)k where b and b are means for m and r items, respectively. To find the variance...C5 , and C12 were treated as known. We find that the standard errors of B1 to B5 are increased drastically by ignorance of C 1 to C5 ; all...ERIC Facilltv-Acquisitlons Davie Hall 013A 4833 Rugby Avenue Chapel Hill, NC 27514 Bethesda, MD 20014 -7- Dr. A. J. Eschenbrenner 1 Dr. John R
Computation of Standard Errors
Dowd, Bryan E; Greene, William H; Norton, Edward C
2014-01-01
Objectives We discuss the problem of computing the standard errors of functions involving estimated parameters and provide the relevant computer code for three different computational approaches using two popular computer packages. Study Design We show how to compute the standard errors of several functions of interest: the predicted value of the dependent variable for a particular subject, and the effect of a change in an explanatory variable on the predicted value of the dependent variable for an individual subject and average effect for a sample of subjects. Empirical Application Using a publicly available dataset, we explain three different methods of computing standard errors: the delta method, Krinsky–Robb, and bootstrapping. We provide computer code for Stata 12 and LIMDEP 10/NLOGIT 5. Conclusions In most applications, choice of the computational method for standard errors of functions of estimated parameters is a matter of convenience. However, when computing standard errors of the sample average of functions that involve both estimated parameters and nonstochastic explanatory variables, it is important to consider the sources of variation in the function's values. PMID:24800304
Ocean wavenumber estimation from wave-resolving time series imagery
Plant, N.G.; Holland, K.T.; Haller, M.C.
2008-01-01
We review several approaches that have been used to estimate ocean surface gravity wavenumbers from wave-resolving remotely sensed image sequences. Two fundamentally different approaches that utilize these data exist. A power spectral density approach identifies wavenumbers where image intensity variance is maximized. Alternatively, a cross-spectral correlation approach identifies wavenumbers where intensity coherence is maximized. We develop a solution to the latter approach based on a tomographic analysis that utilizes a nonlinear inverse method. The solution is tolerant to noise and other forms of sampling deficiency and can be applied to arbitrary sampling patterns, as well as to full-frame imagery. The solution includes error predictions that can be used for data retrieval quality control and for evaluating sample designs. A quantitative analysis of the intrinsic resolution of the method indicates that the cross-spectral correlation fitting improves resolution by a factor of about ten times as compared to the power spectral density fitting approach. The resolution analysis also provides a rule of thumb for nearshore bathymetry retrievals-short-scale cross-shore patterns may be resolved if they are about ten times longer than the average water depth over the pattern. This guidance can be applied to sample design to constrain both the sensor array (image resolution) and the analysis array (tomographic resolution). ?? 2008 IEEE.
Jacob, Benjamin G; Griffith, Daniel A; Muturi, Ephantus J; Caamano, Erick X; Githure, John I; Novak, Robert J
2009-01-01
Background Autoregressive regression coefficients for Anopheles arabiensis aquatic habitat models are usually assessed using global error techniques and are reported as error covariance matrices. A global statistic, however, will summarize error estimates from multiple habitat locations. This makes it difficult to identify where there are clusters of An. arabiensis aquatic habitats of acceptable prediction. It is therefore useful to conduct some form of spatial error analysis to detect clusters of An. arabiensis aquatic habitats based on uncertainty residuals from individual sampled habitats. In this research, a method of error estimation for spatial simulation models was demonstrated using autocorrelation indices and eigenfunction spatial filters to distinguish among the effects of parameter uncertainty on a stochastic simulation of ecological sampled Anopheles aquatic habitat covariates. A test for diagnostic checking error residuals in an An. arabiensis aquatic habitat model may enable intervention efforts targeting productive habitats clusters, based on larval/pupal productivity, by using the asymptotic distribution of parameter estimates from a residual autocovariance matrix. The models considered in this research extends a normal regression analysis previously considered in the literature. Methods Field and remote-sampled data were collected during July 2006 to December 2007 in Karima rice-village complex in Mwea, Kenya. SAS 9.1.4® was used to explore univariate statistics, correlations, distributions, and to generate global autocorrelation statistics from the ecological sampled datasets. A local autocorrelation index was also generated using spatial covariance parameters (i.e., Moran's Indices) in a SAS/GIS® database. The Moran's statistic was decomposed into orthogonal and uncorrelated synthetic map pattern components using a Poisson model with a gamma-distributed mean (i.e. negative binomial regression). The eigenfunction values from the spatial configuration matrices were then used to define expectations for prior distributions using a Markov chain Monte Carlo (MCMC) algorithm. A set of posterior means were defined in WinBUGS 1.4.3®. After the model had converged, samples from the conditional distributions were used to summarize the posterior distribution of the parameters. Thereafter, a spatial residual trend analyses was used to evaluate variance uncertainty propagation in the model using an autocovariance error matrix. Results By specifying coefficient estimates in a Bayesian framework, the covariate number of tillers was found to be a significant predictor, positively associated with An. arabiensis aquatic habitats. The spatial filter models accounted for approximately 19% redundant locational information in the ecological sampled An. arabiensis aquatic habitat data. In the residual error estimation model there was significant positive autocorrelation (i.e., clustering of habitats in geographic space) based on log-transformed larval/pupal data and the sampled covariate depth of habitat. Conclusion An autocorrelation error covariance matrix and a spatial filter analyses can prioritize mosquito control strategies by providing a computationally attractive and feasible description of variance uncertainty estimates for correctly identifying clusters of prolific An. arabiensis aquatic habitats based on larval/pupal productivity. PMID:19772590
Kunz, Cornelia U; Stallard, Nigel; Parsons, Nicholas; Todd, Susan; Friede, Tim
2017-03-01
Regulatory authorities require that the sample size of a confirmatory trial is calculated prior to the start of the trial. However, the sample size quite often depends on parameters that might not be known in advance of the study. Misspecification of these parameters can lead to under- or overestimation of the sample size. Both situations are unfavourable as the first one decreases the power and the latter one leads to a waste of resources. Hence, designs have been suggested that allow a re-assessment of the sample size in an ongoing trial. These methods usually focus on estimating the variance. However, for some methods the performance depends not only on the variance but also on the correlation between measurements. We develop and compare different methods for blinded estimation of the correlation coefficient that are less likely to introduce operational bias when the blinding is maintained. Their performance with respect to bias and standard error is compared to the unblinded estimator. We simulated two different settings: one assuming that all group means are the same and one assuming that different groups have different means. Simulation results show that the naïve (one-sample) estimator is only slightly biased and has a standard error comparable to that of the unblinded estimator. However, if the group means differ, other estimators have better performance depending on the sample size per group and the number of groups. © 2016 The Authors. Biometrical Journal Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
Stallard, Nigel; Parsons, Nicholas; Todd, Susan; Friede, Tim
2016-01-01
Regulatory authorities require that the sample size of a confirmatory trial is calculated prior to the start of the trial. However, the sample size quite often depends on parameters that might not be known in advance of the study. Misspecification of these parameters can lead to under‐ or overestimation of the sample size. Both situations are unfavourable as the first one decreases the power and the latter one leads to a waste of resources. Hence, designs have been suggested that allow a re‐assessment of the sample size in an ongoing trial. These methods usually focus on estimating the variance. However, for some methods the performance depends not only on the variance but also on the correlation between measurements. We develop and compare different methods for blinded estimation of the correlation coefficient that are less likely to introduce operational bias when the blinding is maintained. Their performance with respect to bias and standard error is compared to the unblinded estimator. We simulated two different settings: one assuming that all group means are the same and one assuming that different groups have different means. Simulation results show that the naïve (one‐sample) estimator is only slightly biased and has a standard error comparable to that of the unblinded estimator. However, if the group means differ, other estimators have better performance depending on the sample size per group and the number of groups. PMID:27886393
NASA Astrophysics Data System (ADS)
Maginnis, P. A.; West, M.; Dullerud, G. E.
2016-10-01
We propose an algorithm to accelerate Monte Carlo simulation for a broad class of stochastic processes. Specifically, the class of countable-state, discrete-time Markov chains driven by additive Poisson noise, or lattice discrete-time Markov chains. In particular, this class includes simulation of reaction networks via the tau-leaping algorithm. To produce the speedup, we simulate pairs of fair-draw trajectories that are negatively correlated. Thus, when averaged, these paths produce an unbiased Monte Carlo estimator that has reduced variance and, therefore, reduced error. Numerical results for three example systems included in this work demonstrate two to four orders of magnitude reduction of mean-square error. The numerical examples were chosen to illustrate different application areas and levels of system complexity. The areas are: gene expression (affine state-dependent rates), aerosol particle coagulation with emission and human immunodeficiency virus infection (both with nonlinear state-dependent rates). Our algorithm views the system dynamics as a ;black-box;, i.e., we only require control of pseudorandom number generator inputs. As a result, typical codes can be retrofitted with our algorithm using only minor changes. We prove several analytical results. Among these, we characterize the relationship of covariances between paths in the general nonlinear state-dependent intensity rates case, and we prove variance reduction of mean estimators in the special case of affine intensity rates.
Interpreting Variance Components as Evidence for Reliability and Validity.
ERIC Educational Resources Information Center
Kane, Michael T.
The reliability and validity of measurement is analyzed by a sampling model based on generalizability theory. A model for the relationship between a measurement procedure and an attribute is developed from an analysis of how measurements are used and interpreted in science. The model provides a basis for analyzing the concept of an error of…
Mortazavi, Forough; Mortazavi, Saideh S; Khosrorad, Razieh
2015-09-01
Procrastination is a common behavior which affects different aspects of life. The procrastination assessment scale-student (PASS) evaluates academic procrastination apropos its frequency and reasons. The aims of the present study were to translate, culturally adapt, and validate the Farsi version of the PASS in a sample of Iranian medical students. In this cross-sectional study, the PASS was translated into Farsi through the forward-backward method, and its content validity was thereafter assessed by a panel of 10 experts. The Farsi version of the PASS was subsequently distributed among 423 medical students. The internal reliability of the PASS was assessed using Cronbach's alpha. An exploratory factor analysis (EFA) was conducted on 18 items and then 28 items of the scale to find new models. The construct validity of the scale was assessed using both EFA and confirmatory factor analysis. The predictive validity of the scale was evaluated by calculating the correlation between the academic procrastination scores and the students' average scores in the previous semester. The corresponding reliability of the first and second parts of the scale was 0.781 and 0.861. An EFA on 18 items of the scale found 4 factors which jointly explained 53.2% of variances: The model was marginally acceptable (root mean square error of approximation [RMSEA] =0.098, standardized root mean square residual [SRMR] =0.076, χ(2) /df =4.8, comparative fit index [CFI] =0.83). An EFA on 28 items of the scale found 4 factors which altogether explained 42.62% of variances: The model was acceptable (RMSEA =0.07, SRMR =0.07, χ(2)/df =2.8, incremental fit index =0.90, CFI =0.90). There was a negative correlation between the procrastination scores and the students' average scores (r = -0.131, P =0.02). The Farsi version of the PASS is a valid and reliable tool to measure academic procrastination in Iranian undergraduate medical students.
An investigation of error correcting techniques for OMV and AXAF
NASA Technical Reports Server (NTRS)
Ingels, Frank; Fryer, John
1991-01-01
The original objectives of this project were to build a test system for the NASA 255/223 Reed/Solomon encoding/decoding chip set and circuit board. This test system was then to be interfaced with a convolutional system at MSFC to examine the performance of the concantinated codes. After considerable work, it was discovered that the convolutional system could not function as needed. This report documents the design, construction, and testing of the test apparatus for the R/S chip set. The approach taken was to verify the error correcting behavior of the chip set by injecting known error patterns onto data and observing the results. Error sequences were generated using pseudo-random number generator programs, with Poisson time distribution between errors and Gaussian burst lengths. Sample means, variances, and number of un-correctable errors were calculated for each data set before testing.
Constraining Particle Variation in Lunar Regolith for Simulant Design
NASA Technical Reports Server (NTRS)
Schrader, Christian M.; Rickman, Doug; Stoeser, Douglas; Hoelzer, Hans
2008-01-01
Simulants are used by the lunar engineering community to develop and test technologies for In Situ Resource Utilization (ISRU), excavation and drilling, and for mitigation of hazards to machinery and human health. Working with the United States Geological Survey (USGS), other NASA centers, private industry and academia, Marshall Space Flight Center (MSFC) is leading NASA s lunar regolith simulant program. There are two main efforts: simulant production and simulant evaluation. This work requires a highly detailed understanding of regolith particle type, size, and shape distribution, and of bulk density. The project has developed Figure of Merit (FoM) algorithms to quantitatively compare these characteristics between two materials. The FoM can be used to compare two lunar regolith samples, regolith to simulant, or two parcels of simulant. In work presented here, we use the FoM algorithm to examine the variance of particle type in Apollo 16 highlands regolith core and surface samples. For this analysis we have used internally consistent particle type data for the 90-150 m fraction of Apollo core 64001/64002 from station 4, core 60009/60010 from station 10, and surface samples from various Apollo 16 stations. We calculate mean modal compositions for each core and for the group of surface samples and quantitatively compare samples of each group to its mean as a measurement of within-group variance; we also calculate an FoM for every sample against the mean composition of 64001/64002. This gives variation with depth at two locations and between Apollo 16 stations. Of the tested groups, core 60009/60010 has the highest internal variance with an average FoM score of 0.76 and core 64001/64002 has the lowest with an average FoM of 0.92. The surface samples have a low but intermediate internal variance with an average FoM of 0.79. FoM s calculated against the 64001/64002 mean reference composition range from 0.79-0.97 for 64001/64002, from 0.41-0.91 for 60009/60010, and from 0.54-0.93 for the surface samples. Six samples fall below 0.70, and they are also the least mature (i.e., have the lowest I(sub s)/FeO). Because agglutinates are the dominant particle type and the agglutinate population increases with sample maturity (I(sub s)/FeO), the maturity of the sample relative to the reference is a prime determinant of the particle type FoM score within these highland samples.
Performance Analysis of Classification Methods for Indoor Localization in Vlc Networks
NASA Astrophysics Data System (ADS)
Sánchez-Rodríguez, D.; Alonso-González, I.; Sánchez-Medina, J.; Ley-Bosch, C.; Díaz-Vilariño, L.
2017-09-01
Indoor localization has gained considerable attention over the past decade because of the emergence of numerous location-aware services. Research works have been proposed on solving this problem by using wireless networks. Nevertheless, there is still much room for improvement in the quality of the proposed classification models. In the last years, the emergence of Visible Light Communication (VLC) brings a brand new approach to high quality indoor positioning. Among its advantages, this new technology is immune to electromagnetic interference and has the advantage of having a smaller variance of received signal power compared to RF based technologies. In this paper, a performance analysis of seventeen machine leaning classifiers for indoor localization in VLC networks is carried out. The analysis is accomplished in terms of accuracy, average distance error, computational cost, training size, precision and recall measurements. Results show that most of classifiers harvest an accuracy above 90 %. The best tested classifier yielded a 99.0 % accuracy, with an average error distance of 0.3 centimetres.
Characterization of impulse noise and analysis of its effect upon correlation receivers
NASA Technical Reports Server (NTRS)
Houts, R. C.; Moore, J. D.
1971-01-01
A noise model is formulated to describe the impulse noise in many digital systems. A simplified model, which assumes that each noise burst contains a randomly weighted version of the same basic waveform, is used to derive the performance equations for a correlation receiver. The expected number of bit errors per noise burst is expressed as a function of the average signal energy, signal-set correlation coefficient, bit time, noise-weighting-factor variance and probability density function, and a time range function which depends on the crosscorrelation of the signal-set basis functions and the noise waveform. A procedure is established for extending the results for the simplified noise model to the general model. Unlike the performance results for Gaussian noise, it is shown that for impulse noise the error performance is affected by the choice of signal-set basis functions and that Orthogonal signaling is not equivalent to On-Off signaling with the same average energy.
Honest Importance Sampling with Multiple Markov Chains
Tan, Aixin; Doss, Hani; Hobert, James P.
2017-01-01
Importance sampling is a classical Monte Carlo technique in which a random sample from one probability density, π1, is used to estimate an expectation with respect to another, π. The importance sampling estimator is strongly consistent and, as long as two simple moment conditions are satisfied, it obeys a central limit theorem (CLT). Moreover, there is a simple consistent estimator for the asymptotic variance in the CLT, which makes for routine computation of standard errors. Importance sampling can also be used in the Markov chain Monte Carlo (MCMC) context. Indeed, if the random sample from π1 is replaced by a Harris ergodic Markov chain with invariant density π1, then the resulting estimator remains strongly consistent. There is a price to be paid however, as the computation of standard errors becomes more complicated. First, the two simple moment conditions that guarantee a CLT in the iid case are not enough in the MCMC context. Second, even when a CLT does hold, the asymptotic variance has a complex form and is difficult to estimate consistently. In this paper, we explain how to use regenerative simulation to overcome these problems. Actually, we consider a more general set up, where we assume that Markov chain samples from several probability densities, π1, …, πk, are available. We construct multiple-chain importance sampling estimators for which we obtain a CLT based on regeneration. We show that if the Markov chains converge to their respective target distributions at a geometric rate, then under moment conditions similar to those required in the iid case, the MCMC-based importance sampling estimator obeys a CLT. Furthermore, because the CLT is based on a regenerative process, there is a simple consistent estimator of the asymptotic variance. We illustrate the method with two applications in Bayesian sensitivity analysis. The first concerns one-way random effects models under different priors. The second involves Bayesian variable selection in linear regression, and for this application, importance sampling based on multiple chains enables an empirical Bayes approach to variable selection. PMID:28701855
Honest Importance Sampling with Multiple Markov Chains.
Tan, Aixin; Doss, Hani; Hobert, James P
2015-01-01
Importance sampling is a classical Monte Carlo technique in which a random sample from one probability density, π 1 , is used to estimate an expectation with respect to another, π . The importance sampling estimator is strongly consistent and, as long as two simple moment conditions are satisfied, it obeys a central limit theorem (CLT). Moreover, there is a simple consistent estimator for the asymptotic variance in the CLT, which makes for routine computation of standard errors. Importance sampling can also be used in the Markov chain Monte Carlo (MCMC) context. Indeed, if the random sample from π 1 is replaced by a Harris ergodic Markov chain with invariant density π 1 , then the resulting estimator remains strongly consistent. There is a price to be paid however, as the computation of standard errors becomes more complicated. First, the two simple moment conditions that guarantee a CLT in the iid case are not enough in the MCMC context. Second, even when a CLT does hold, the asymptotic variance has a complex form and is difficult to estimate consistently. In this paper, we explain how to use regenerative simulation to overcome these problems. Actually, we consider a more general set up, where we assume that Markov chain samples from several probability densities, π 1 , …, π k , are available. We construct multiple-chain importance sampling estimators for which we obtain a CLT based on regeneration. We show that if the Markov chains converge to their respective target distributions at a geometric rate, then under moment conditions similar to those required in the iid case, the MCMC-based importance sampling estimator obeys a CLT. Furthermore, because the CLT is based on a regenerative process, there is a simple consistent estimator of the asymptotic variance. We illustrate the method with two applications in Bayesian sensitivity analysis. The first concerns one-way random effects models under different priors. The second involves Bayesian variable selection in linear regression, and for this application, importance sampling based on multiple chains enables an empirical Bayes approach to variable selection.
Monte Carlo isotopic inventory analysis for complex nuclear systems
NASA Astrophysics Data System (ADS)
Phruksarojanakun, Phiphat
Monte Carlo Inventory Simulation Engine (MCise) is a newly developed method for calculating isotopic inventory of materials. It offers the promise of modeling materials with complex processes and irradiation histories, which pose challenges for current, deterministic tools, and has strong analogies to Monte Carlo (MC) neutral particle transport. The analog method, including considerations for simple, complex and loop flows, is fully developed. In addition, six variance reduction tools provide unique capabilities of MCise to improve statistical precision of MC simulations. Forced Reaction forces an atom to undergo a desired number of reactions in a given irradiation environment. Biased Reaction Branching primarily focuses on improving statistical results of the isotopes that are produced from rare reaction pathways. Biased Source Sampling aims at increasing frequencies of sampling rare initial isotopes as the starting particles. Reaction Path Splitting increases the population by splitting the atom at each reaction point, creating one new atom for each decay or transmutation product. Delta Tracking is recommended for high-frequency pulsing to reduce the computing time. Lastly, Weight Window is introduced as a strategy to decrease large deviations of weight due to the uses of variance reduction techniques. A figure of merit is necessary to compare the efficiency of different variance reduction techniques. A number of possibilities for figure of merit are explored, two of which are robust and subsequently used. One is based on the relative error of a known target isotope (1/R 2T) and the other on the overall detection limit corrected by the relative error (1/DkR 2T). An automated Adaptive Variance-reduction Adjustment (AVA) tool is developed to iteratively define parameters for some variance reduction techniques in a problem with a target isotope. Sample problems demonstrate that AVA improves both precision and accuracy of a target result in an efficient manner. Potential applications of MCise include molten salt fueled reactors and liquid breeders in fusion blankets. As an example, the inventory analysis of a liquid actinide fuel in the In-Zinerator, a sub-critical power reactor driven by a fusion source, is examined. The result reassures MCise as a reliable tool for inventory analysis of complex nuclear systems.
Speeding up Coarse Point Cloud Registration by Threshold-Independent Baysac Match Selection
NASA Astrophysics Data System (ADS)
Kang, Z.; Lindenbergh, R.; Pu, S.
2016-06-01
This paper presents an algorithm for the automatic registration of terrestrial point clouds by match selection using an efficiently conditional sampling method -- threshold-independent BaySAC (BAYes SAmpling Consensus) and employs the error metric of average point-to-surface residual to reduce the random measurement error and then approach the real registration error. BaySAC and other basic sampling algorithms usually need to artificially determine a threshold by which inlier points are identified, which leads to a threshold-dependent verification process. Therefore, we applied the LMedS method to construct the cost function that is used to determine the optimum model to reduce the influence of human factors and improve the robustness of the model estimate. Point-to-point and point-to-surface error metrics are most commonly used. However, point-to-point error in general consists of at least two components, random measurement error and systematic error as a result of a remaining error in the found rigid body transformation. Thus we employ the measure of the average point-to-surface residual to evaluate the registration accuracy. The proposed approaches, together with a traditional RANSAC approach, are tested on four data sets acquired by three different scanners in terms of their computational efficiency and quality of the final registration. The registration results show the st.dev of the average point-to-surface residuals is reduced from 1.4 cm (plain RANSAC) to 0.5 cm (threshold-independent BaySAC). The results also show that, compared to the performance of RANSAC, our BaySAC strategies lead to less iterations and cheaper computational cost when the hypothesis set is contaminated with more outliers.
Fiddler, W; Pensabene, J W; Gates, R A; Phillips, J G
1984-01-01
A dry column method for isolating N-nitrosopyrrolidine (NPYR) from fried, cure-pumped bacon and detection by gas chromatography-thermal energy analyzer (TEA) was studied collaboratively. Testing the results obtained from 11 collaborators for homogeneous variances among samples resulted in splitting the nonzero samples into 2 groups of sample levels, each with similar variances. Outlying results were identified by AOAC-recommended procedures, and laboratories having outliers within a group were excluded. Results from the 9 collaborators remaining in the low group yielded coefficients of variation (CV) of 6.00% and 7.47% for repeatability and reproducibility, respectively, and the 8 collaborators remaining in the high group yielded CV values of 5.64% and 13.72%, respectively. An 85.2% overall average recovery of the N-nitrosoazetidine internal standard was obtained with an average laboratory CV of 10.5%. The method has been adopted official first action as an alternative to the mineral oil distillation-TEA screening procedure.
Blokland, Gabriëlla A M; Mesholam-Gately, Raquelle I; Toulopoulou, Timothea; Del Re, Elisabetta C; Lam, Max; DeLisi, Lynn E; Donohoe, Gary; Walters, James T R; Seidman, Larry J; Petryshen, Tracey L
2017-07-01
Schizophrenia is characterized by neuropsychological deficits across many cognitive domains. Cognitive phenotypes with high heritability and genetic overlap with schizophrenia liability can help elucidate the mechanisms leading from genes to psychopathology. We performed a meta-analysis of 170 published twin and family heritability studies of >800 000 nonpsychiatric and schizophrenia subjects to accurately estimate heritability across many neuropsychological tests and cognitive domains. The proportion of total variance of each phenotype due to additive genetic effects (A), shared environment (C), and unshared environment and error (E), was calculated by averaging A, C, and E estimates across studies and weighting by sample size. Heritability ranged across phenotypes, likely due to differences in genetic and environmental effects, with the highest heritability for General Cognitive Ability (32%-67%), Verbal Ability (43%-72%), Visuospatial Ability (20%-80%), and Attention/Processing Speed (28%-74%), while the lowest heritability was observed for Executive Function (20%-40%). These results confirm that many cognitive phenotypes are under strong genetic influences. Heritability estimates were comparable in nonpsychiatric and schizophrenia samples, suggesting that environmental factors and illness-related moderators (eg, medication) do not substantially decrease heritability in schizophrenia samples, and that genetic studies in schizophrenia samples are informative for elucidating the genetic basis of cognitive deficits. Substantial genetic overlap between cognitive phenotypes and schizophrenia liability (average rg = -.58) in twin studies supports partially shared genetic etiology. It will be important to conduct comparative studies in well-powered samples to determine whether the same or different genes and genetic variants influence cognition in schizophrenia patients and the general population. © The Author 2016. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved. For permissions, please email: journals.permissions@oup.com.
Design and analysis of three-arm trials with negative binomially distributed endpoints.
Mütze, Tobias; Munk, Axel; Friede, Tim
2016-02-20
A three-arm clinical trial design with an experimental treatment, an active control, and a placebo control, commonly referred to as the gold standard design, enables testing of non-inferiority or superiority of the experimental treatment compared with the active control. In this paper, we propose methods for designing and analyzing three-arm trials with negative binomially distributed endpoints. In particular, we develop a Wald-type test with a restricted maximum-likelihood variance estimator for testing non-inferiority or superiority. For this test, sample size and power formulas as well as optimal sample size allocations will be derived. The performance of the proposed test will be assessed in an extensive simulation study with regard to type I error rate, power, sample size, and sample size allocation. For the purpose of comparison, Wald-type statistics with a sample variance estimator and an unrestricted maximum-likelihood estimator are included in the simulation study. We found that the proposed Wald-type test with a restricted variance estimator performed well across the considered scenarios and is therefore recommended for application in clinical trials. The methods proposed are motivated and illustrated by a recent clinical trial in multiple sclerosis. The R package ThreeArmedTrials, which implements the methods discussed in this paper, is available on CRAN. Copyright © 2015 John Wiley & Sons, Ltd.
Non-linear matter power spectrum covariance matrix errors and cosmological parameter uncertainties
NASA Astrophysics Data System (ADS)
Blot, L.; Corasaniti, P. S.; Amendola, L.; Kitching, T. D.
2016-06-01
The covariance of the matter power spectrum is a key element of the analysis of galaxy clustering data. Independent realizations of observational measurements can be used to sample the covariance, nevertheless statistical sampling errors will propagate into the cosmological parameter inference potentially limiting the capabilities of the upcoming generation of galaxy surveys. The impact of these errors as function of the number of realizations has been previously evaluated for Gaussian distributed data. However, non-linearities in the late-time clustering of matter cause departures from Gaussian statistics. Here, we address the impact of non-Gaussian errors on the sample covariance and precision matrix errors using a large ensemble of N-body simulations. In the range of modes where finite volume effects are negligible (0.1 ≲ k [h Mpc-1] ≲ 1.2), we find deviations of the variance of the sample covariance with respect to Gaussian predictions above ˜10 per cent at k > 0.3 h Mpc-1. Over the entire range these reduce to about ˜5 per cent for the precision matrix. Finally, we perform a Fisher analysis to estimate the effect of covariance errors on the cosmological parameter constraints. In particular, assuming Euclid-like survey characteristics we find that a number of independent realizations larger than 5000 is necessary to reduce the contribution of sampling errors to the cosmological parameter uncertainties at subpercent level. We also show that restricting the analysis to large scales k ≲ 0.2 h Mpc-1 results in a considerable loss in constraining power, while using the linear covariance to include smaller scales leads to an underestimation of the errors on the cosmological parameters.
Satellites for the study of ocean primary productivity
NASA Technical Reports Server (NTRS)
Smith, R. C.; Baker, K. S.
1983-01-01
The use of remote sensing techniques for obtaining estimates of global marine primary productivity is examined. It is shown that remote sensing and multiplatform (ship, aircraft, and satellite) sampling strategies can be used to significantly lower the variance in estimates of phytoplankton abundance and of population growth rates from the values obtained using the C-14 method. It is noted that multiplatform sampling strategies are essential to assess the mean and variance of phytoplankton biomass on a regional or on a global basis. The relative errors associated with shipboard and satellite estimates of phytoplankton biomass and primary productivity, as well as the increased statistical accuracy possible from the utilization of contemporaneous data from both sampling platforms, are examined. It is shown to be possible to follow changes in biomass and the distribution patterns of biomass as a function of time with the use of satellite imagery.
Pérennes, Maud; Carde, Axel; Nicolas, Xavier; Dolz, Manuel; Bihannic, René; Grimont, Pauline; Chapot, Thierry; Granier, Hervé
2012-03-01
An inaccurate medication history may prevent the discovery of a pre-admission iatrogenic event or lead to interrupted drug therapy during hospitalization. Medication reconciliation is a process that ensures the transfer of medication information at admission to the hospital. The aims of this prospective study were to evaluate the interest in clinical practice of this concept and the resources needed for its implementation. We chose to include patients aged 65 years or over admitted in the internal medicine unit between June and October 2010. We obtained an accurate list of each patient's home medications. This list was then compared with medication orders. All medication variances were classified as intended or unintended. An internist and a pharmacist classified the clinical importance of each unintended variance. Sixty-one patients (mean age: 78 ± 7.4 years) were included in our study. We identified 38 unintended discrepancies. The average number of unintended discrepancies was 0.62 per patient. Twenty-five patients (41%) had one or more unintended discrepancies at admission. The contact with the community pharmacist permitted us to identify 21 (55%) unintended discrepancies. The most common errors were the omission of a regularly used medication (76%) and an incorrect dosage (16%). Our intervention resulted in order changes by the physician for 30 (79%) unintended discrepancies. Fifty percent of the unintended variances were judged by the internist and 76% by the pharmacist to be clinically significant. The admission to the hospital is a critical transition point for the continuity of care in medication management. Medication reconciliation can identify and resolve errors due to inaccurate medication histories. Copyright © 2011 Elsevier Masson SAS. All rights reserved.
Bohmanova, J; Miglior, F; Jamrozik, J; Misztal, I; Sullivan, P G
2008-09-01
A random regression model with both random and fixed regressions fitted by Legendre polynomials of order 4 was compared with 3 alternative models fitting linear splines with 4, 5, or 6 knots. The effects common for all models were a herd-test-date effect, fixed regressions on days in milk (DIM) nested within region-age-season of calving class, and random regressions for additive genetic and permanent environmental effects. Data were test-day milk, fat and protein yields, and SCS recorded from 5 to 365 DIM during the first 3 lactations of Canadian Holstein cows. A random sample of 50 herds consisting of 96,756 test-day records was generated to estimate variance components within a Bayesian framework via Gibbs sampling. Two sets of genetic evaluations were subsequently carried out to investigate performance of the 4 models. Models were compared by graphical inspection of variance functions, goodness of fit, error of prediction of breeding values, and stability of estimated breeding values. Models with splines gave lower estimates of variances at extremes of lactations than the model with Legendre polynomials. Differences among models in goodness of fit measured by percentages of squared bias, correlations between predicted and observed records, and residual variances were small. The deviance information criterion favored the spline model with 6 knots. Smaller error of prediction and higher stability of estimated breeding values were achieved by using spline models with 5 and 6 knots compared with the model with Legendre polynomials. In general, the spline model with 6 knots had the best overall performance based upon the considered model comparison criteria.
A BASIS FOR MODIFYING THE TANK 12 COMPOSITE SAMPLING DESIGN
DOE Office of Scientific and Technical Information (OSTI.GOV)
Shine, G.
The SRR sampling campaign to obtain residual solids material from the Savannah River Site (SRS) Tank Farm Tank 12 primary vessel resulted in obtaining appreciable material in all 6 planned source samples from the mound strata but only in 5 of the 6 planned source samples from the floor stratum. Consequently, the design of the compositing scheme presented in the Tank 12 Sampling and Analysis Plan, Pavletich (2014a), must be revised. Analytical Development of SRNL statistically evaluated the sampling uncertainty associated with using various compositing arrays and splitting one or more samples for compositing. The variance of the simple meanmore » of composite sample concentrations is a reasonable standard to investigate the impact of the following sampling options. Composite Sample Design Option (a). Assign only 1 source sample from the floor stratum and 1 source sample from each of the mound strata to each of the composite samples. Each source sample contributes material to only 1 composite sample. Two source samples from the floor stratum would not be used. Composite Sample Design Option (b). Assign 2 source samples from the floor stratum and 1 source sample from each of the mound strata to each composite sample. This infers that one source sample from the floor must be used twice, with 2 composite samples sharing material from this particular source sample. All five source samples from the floor would be used. Composite Sample Design Option (c). Assign 3 source samples from the floor stratum and 1 source sample from each of the mound strata to each composite sample. This infers that several of the source samples from the floor stratum must be assigned to more than one composite sample. All 5 source samples from the floor would be used. Using fewer than 12 source samples will increase the sampling variability over that of the Basic Composite Sample Design, Pavletich (2013). Considering the impact to the variance of the simple mean of the composite sample concentrations, the recommendation is to construct each sample composite using four or five source samples. Although the variance using 5 source samples per composite sample (Composite Sample Design Option (c)) was slightly less than the variance using 4 source samples per composite sample (Composite Sample Design Option (b)), there is no practical difference between those variances. This does not consider that the measurement error variance, which is the same for all composite sample design options considered in this report, will further dilute any differences. Composite Sample Design Option (a) had the largest variance for the mean concentration in the three composite samples and should be avoided. These results are consistent with Pavletich (2014b) which utilizes a low elevation and a high elevation mound source sample and two floor source samples for each composite sample. Utilizing the four source samples per composite design, Pavletich (2014b) utilizes aliquots of Floor Sample 4 for two composite samples.« less
NASA Technical Reports Server (NTRS)
Melbourne, William G.
1986-01-01
In double differencing a regression system obtained from concurrent Global Positioning System (GPS) observation sequences, one either undersamples the system to avoid introducing colored measurement statistics, or one fully samples the system incurring the resulting non-diagonal covariance matrix for the differenced measurement errors. A suboptimal estimation result will be obtained in the undersampling case and will also be obtained in the fully sampled case unless the color noise statistics are taken into account. The latter approach requires a least squares weighting matrix derived from inversion of a non-diagonal covariance matrix for the differenced measurement errors instead of inversion of the customary diagonal one associated with white noise processes. Presented is the so-called fully redundant double differencing algorithm for generating a weighted double differenced regression system that yields equivalent estimation results, but features for certain cases a diagonal weighting matrix even though the differenced measurement error statistics are highly colored.
Guan, Yongtao; Li, Yehua; Sinha, Rajita
2011-01-01
In a cocaine dependence treatment study, we use linear and nonlinear regression models to model posttreatment cocaine craving scores and first cocaine relapse time. A subset of the covariates are summary statistics derived from baseline daily cocaine use trajectories, such as baseline cocaine use frequency and average daily use amount. These summary statistics are subject to estimation error and can therefore cause biased estimators for the regression coefficients. Unlike classical measurement error problems, the error we encounter here is heteroscedastic with an unknown distribution, and there are no replicates for the error-prone variables or instrumental variables. We propose two robust methods to correct for the bias: a computationally efficient method-of-moments-based method for linear regression models and a subsampling extrapolation method that is generally applicable to both linear and nonlinear regression models. Simulations and an application to the cocaine dependence treatment data are used to illustrate the efficacy of the proposed methods. Asymptotic theory and variance estimation for the proposed subsampling extrapolation method and some additional simulation results are described in the online supplementary material. PMID:21984854
Sampling design for spatially distributed hydrogeologic and environmental processes
Christakos, G.; Olea, R.A.
1992-01-01
A methodology for the design of sampling networks over space is proposed. The methodology is based on spatial random field representations of nonhomogeneous natural processes, and on optimal spatial estimation techniques. One of the most important results of random field theory for physical sciences is its rationalization of correlations in spatial variability of natural processes. This correlation is extremely important both for interpreting spatially distributed observations and for predictive performance. The extent of site sampling and the types of data to be collected will depend on the relationship of subsurface variability to predictive uncertainty. While hypothesis formulation and initial identification of spatial variability characteristics are based on scientific understanding (such as knowledge of the physics of the underlying phenomena, geological interpretations, intuition and experience), the support offered by field data is statistically modelled. This model is not limited by the geometric nature of sampling and covers a wide range in subsurface uncertainties. A factorization scheme of the sampling error variance is derived, which possesses certain atttactive properties allowing significant savings in computations. By means of this scheme, a practical sampling design procedure providing suitable indices of the sampling error variance is established. These indices can be used by way of multiobjective decision criteria to obtain the best sampling strategy. Neither the actual implementation of the in-situ sampling nor the solution of the large spatial estimation systems of equations are necessary. The required values of the accuracy parameters involved in the network design are derived using reference charts (readily available for various combinations of data configurations and spatial variability parameters) and certain simple yet accurate analytical formulas. Insight is gained by applying the proposed sampling procedure to realistic examples related to sampling problems in two dimensions. ?? 1992.
Zhang, Zhenwei; VanSwearingen, Jessie; Brach, Jennifer S.; Perera, Subashan
2016-01-01
Human gait is a complex interaction of many nonlinear systems and stride intervals exhibit self-similarity over long time scales that can be modeled as a fractal process. The scaling exponent represents the fractal degree and can be interpreted as a biomarker of relative diseases. The previous study showed that the average wavelet method provides the most accurate results to estimate this scaling exponent when applied to stride interval time series. The purpose of this paper is to determine the most suitable mother wavelet for the average wavelet method. This paper presents a comparative numerical analysis of sixteen mother wavelets using simulated and real fractal signals. Simulated fractal signals were generated under varying signal lengths and scaling exponents that indicate a range of physiologically conceivable fractal signals. The five candidates were chosen due to their good performance on the mean square error test for both short and long signals. Next, we comparatively analyzed these five mother wavelets for physiologically relevant stride time series lengths. Our analysis showed that the symlet 2 mother wavelet provides a low mean square error and low variance for long time intervals and relatively low errors for short signal lengths. It can be considered as the most suitable mother function without the burden of considering the signal length. PMID:27960102
Trends in Classroom Observation Scores
Lockwood, J. R.; McCaffrey, Daniel F.
2014-01-01
Observations and ratings of classroom teaching and interactions collected over time are susceptible to trends in both the quality of instruction and rater behavior. These trends have potential implications for inferences about teaching and for study design. We use scores on the Classroom Assessment Scoring System–Secondary (CLASS-S) protocol from 458 middle school teachers over a 2-year period to study changes over time in (a) the average quality of teaching for the population of teachers, (b) the average severity of the population of raters, and (c) the severity of individual raters. To obtain these estimates and assess them in the context of other factors that contribute to the variability in scores, we develop an augmented G study model that is broadly applicable for modeling sources of variability in classroom observation ratings data collected over time. In our data, we found that trends in teaching quality were small. Rater drift was very large during raters’ initial days of observation and persisted throughout nearly 2 years of scoring. Raters did not converge to a common level of severity; using our model we estimate that variability among raters actually increases over the course of the study. Variance decompositions based on the model find that trends are a modest source of variance relative to overall rater effects, rater errors on specific lessons, and residual error. The discussion provides possible explanations for trends and rater divergence as well as implications for designs collecting ratings over time. PMID:29795823
Trends in Classroom Observation Scores.
Casabianca, Jodi M; Lockwood, J R; McCaffrey, Daniel F
2015-04-01
Observations and ratings of classroom teaching and interactions collected over time are susceptible to trends in both the quality of instruction and rater behavior. These trends have potential implications for inferences about teaching and for study design. We use scores on the Classroom Assessment Scoring System-Secondary (CLASS-S) protocol from 458 middle school teachers over a 2-year period to study changes over time in (a) the average quality of teaching for the population of teachers, (b) the average severity of the population of raters, and (c) the severity of individual raters. To obtain these estimates and assess them in the context of other factors that contribute to the variability in scores, we develop an augmented G study model that is broadly applicable for modeling sources of variability in classroom observation ratings data collected over time. In our data, we found that trends in teaching quality were small. Rater drift was very large during raters' initial days of observation and persisted throughout nearly 2 years of scoring. Raters did not converge to a common level of severity; using our model we estimate that variability among raters actually increases over the course of the study. Variance decompositions based on the model find that trends are a modest source of variance relative to overall rater effects, rater errors on specific lessons, and residual error. The discussion provides possible explanations for trends and rater divergence as well as implications for designs collecting ratings over time.
Jager, Justin; Bornstein, Marc H; Putnick, Diane L; Hendricks, Charlene
2012-06-01
Using the McMaster Family Assessment Device (Epstein, Baldwin, & Bishop, 1983) and incorporating the perspectives of adolescent, mother, and father, this study examined each family member's "unique perspective" or nonshared, idiosyncratic view of the family. We used a modified multitrait-multimethod confirmatory factor analysis that (a) isolated for each family member's 6 reports of family dysfunction the nonshared variance (a combination of variance idiosyncratic to the individual and measurement error) from variance shared by 1 or more family members and (b) extracted common variance across each family member's set of nonshared variances. The sample included 128 families from a U.S. East Coast metropolitan area. Each family member's unique perspective generalized across his or her different reports of family dysfunction and accounted for a sizable proportion of his or her own variance in reports of family dysfunction. In addition, after holding level of dysfunction constant across families and controlling for a family's shared variance (agreement regarding family dysfunction), each family member's unique perspective was associated with his or her own adjustment. Future applications and competing alternatives for what these "unique perspectives" reflect about the family are discussed. PsycINFO Database Record (c) 2012 APA, all rights reserved.
Jager, Justin; Bornstein, Marc H.; Diane, L. Putnick; Hendricks, Charlene
2012-01-01
Using the Family Assessment Device (FAD; Epstein, Baldwin, & Bishop, 1983) and incorporating the perspectives of adolescent, mother, and father, this study examined each family member's “unique perspective” or non-shared, idiosyncratic view of the family. To do so we used a modified multitrait-multimethod confirmatory factor analysis that (1) isolated for each family member's six reports of family dysfunction the non-shared variance (a combination of variance idiosyncratic to the individual and measurement error) from variance shared by one or more family members and (2) extracted common variance across each family member's set of non-shared variances. The sample included 128 families from a U.S. East Coast metropolitan area. Each family member's unique perspective generalized across his or her different reports of family dysfunction and accounted for a sizable proportion of his or her own variance in reports of family dysfunction. Additionally, after holding level of dysfunction constant across families and controlling for a family's shared variance (agreement regarding family dysfunction), each family member's unique perspective was associated with his or her own adjustment. Future applications and competing alternatives for what these “unique perspectives” reflect about the family are discussed. PMID:22545933
Variance reduction for Fokker–Planck based particle Monte Carlo schemes
DOE Office of Scientific and Technical Information (OSTI.GOV)
Gorji, M. Hossein, E-mail: gorjih@ifd.mavt.ethz.ch; Andric, Nemanja; Jenny, Patrick
Recently, Fokker–Planck based particle Monte Carlo schemes have been proposed and evaluated for simulations of rarefied gas flows [1–3]. In this paper, the variance reduction for particle Monte Carlo simulations based on the Fokker–Planck model is considered. First, deviational based schemes were derived and reviewed, and it is shown that these deviational methods are not appropriate for practical Fokker–Planck based rarefied gas flow simulations. This is due to the fact that the deviational schemes considered in this study lead either to instabilities in the case of two-weight methods or to large statistical errors if the direct sampling method is applied.more » Motivated by this conclusion, we developed a novel scheme based on correlated stochastic processes. The main idea here is to synthesize an additional stochastic process with a known solution, which is simultaneously solved together with the main one. By correlating the two processes, the statistical errors can dramatically be reduced; especially for low Mach numbers. To assess the methods, homogeneous relaxation, planar Couette and lid-driven cavity flows were considered. For these test cases, it could be demonstrated that variance reduction based on parallel processes is very robust and effective.« less
Chiu, Mei Choi; Pun, Chi Seng; Wong, Hoi Ying
2017-08-01
Investors interested in the global financial market must analyze financial securities internationally. Making an optimal global investment decision involves processing a huge amount of data for a high-dimensional portfolio. This article investigates the big data challenges of two mean-variance optimal portfolios: continuous-time precommitment and constant-rebalancing strategies. We show that both optimized portfolios implemented with the traditional sample estimates converge to the worst performing portfolio when the portfolio size becomes large. The crux of the problem is the estimation error accumulated from the huge dimension of stock data. We then propose a linear programming optimal (LPO) portfolio framework, which applies a constrained ℓ 1 minimization to the theoretical optimal control to mitigate the risk associated with the dimensionality issue. The resulting portfolio becomes a sparse portfolio that selects stocks with a data-driven procedure and hence offers a stable mean-variance portfolio in practice. When the number of observations becomes large, the LPO portfolio converges to the oracle optimal portfolio, which is free of estimation error, even though the number of stocks grows faster than the number of observations. Our numerical and empirical studies demonstrate the superiority of the proposed approach. © 2017 Society for Risk Analysis.
flowVS: channel-specific variance stabilization in flow cytometry.
Azad, Ariful; Rajwa, Bartek; Pothen, Alex
2016-07-28
Comparing phenotypes of heterogeneous cell populations from multiple biological conditions is at the heart of scientific discovery based on flow cytometry (FC). When the biological signal is measured by the average expression of a biomarker, standard statistical methods require that variance be approximately stabilized in populations to be compared. Since the mean and variance of a cell population are often correlated in fluorescence-based FC measurements, a preprocessing step is needed to stabilize the within-population variances. We present a variance-stabilization algorithm, called flowVS, that removes the mean-variance correlations from cell populations identified in each fluorescence channel. flowVS transforms each channel from all samples of a data set by the inverse hyperbolic sine (asinh) transformation. For each channel, the parameters of the transformation are optimally selected by Bartlett's likelihood-ratio test so that the populations attain homogeneous variances. The optimum parameters are then used to transform the corresponding channels in every sample. flowVS is therefore an explicit variance-stabilization method that stabilizes within-population variances in each channel by evaluating the homoskedasticity of clusters with a likelihood-ratio test. With two publicly available datasets, we show that flowVS removes the mean-variance dependence from raw FC data and makes the within-population variance relatively homogeneous. We demonstrate that alternative transformation techniques such as flowTrans, flowScape, logicle, and FCSTrans might not stabilize variance. Besides flow cytometry, flowVS can also be applied to stabilize variance in microarray data. With a publicly available data set we demonstrate that flowVS performs as well as the VSN software, a state-of-the-art approach developed for microarrays. The homogeneity of variance in cell populations across FC samples is desirable when extracting features uniformly and comparing cell populations with different levels of marker expressions. The newly developed flowVS algorithm solves the variance-stabilization problem in FC and microarrays by optimally transforming data with the help of Bartlett's likelihood-ratio test. On two publicly available FC datasets, flowVS stabilizes within-population variances more evenly than the available transformation and normalization techniques. flowVS-based variance stabilization can help in performing comparison and alignment of phenotypically identical cell populations across different samples. flowVS and the datasets used in this paper are publicly available in Bioconductor.
Performance of Language-Coordinated Collective Systems: A Study of Wine Recognition and Description
Zubek, Julian; Denkiewicz, Michał; Dębska, Agnieszka; Radkowska, Alicja; Komorowska-Mach, Joanna; Litwin, Piotr; Stępień, Magdalena; Kucińska, Adrianna; Sitarska, Ewa; Komorowska, Krystyna; Fusaroli, Riccardo; Tylén, Kristian; Rączaszek-Leonardi, Joanna
2016-01-01
Most of our perceptions of and engagements with the world are shaped by our immersion in social interactions, cultural traditions, tools and linguistic categories. In this study we experimentally investigate the impact of two types of language-based coordination on the recognition and description of complex sensory stimuli: that of red wine. Participants were asked to taste, remember and successively recognize samples of wines within a larger set in a two-by-two experimental design: (1) either individually or in pairs, and (2) with or without the support of a sommelier card—a cultural linguistic tool designed for wine description. Both effectiveness of recognition and the kinds of errors in the four conditions were analyzed. While our experimental manipulations did not impact recognition accuracy, bias-variance decomposition of error revealed non-trivial differences in how participants solved the task. Pairs generally displayed reduced bias and increased variance compared to individuals, however the variance dropped significantly when they used the sommelier card. The effect of sommelier card reducing the variance was observed only in pairs, individuals did not seem to benefit from the cultural linguistic tool. Analysis of descriptions generated with the aid of sommelier cards shows that pairs were more coherent and discriminative than individuals. The findings are discussed in terms of global properties and dynamics of collective systems when constrained by different types of cultural practices. PMID:27729875
Daboul, Amro; Ivanovska, Tatyana; Bülow, Robin; Biffar, Reiner; Cardini, Andrea
2018-01-01
Using 3D anatomical landmarks from adult human head MRIs, we assessed the magnitude of inter-operator differences in Procrustes-based geometric morphometric analyses. An in depth analysis of both absolute and relative error was performed in a subsample of individuals with replicated digitization by three different operators. The effect of inter-operator differences was also explored in a large sample of more than 900 individuals. Although absolute error was not unusual for MRI measurements, including bone landmarks, shape was particularly affected by differences among operators, with up to more than 30% of sample variation accounted for by this type of error. The magnitude of the bias was such that it dominated the main pattern of bone and total (all landmarks included) shape variation, largely surpassing the effect of sex differences between hundreds of men and women. In contrast, however, we found higher reproducibility in soft-tissue nasal landmarks, despite relatively larger errors in estimates of nasal size. Our study exemplifies the assessment of measurement error using geometric morphometrics on landmarks from MRIs and stresses the importance of relating it to total sample variance within the specific methodological framework being used. In summary, precise landmarks may not necessarily imply negligible errors, especially in shape data; indeed, size and shape may be differentially impacted by measurement error and different types of landmarks may have relatively larger or smaller errors. Importantly, and consistently with other recent studies using geometric morphometrics on digital images (which, however, were not specific to MRI data), this study showed that inter-operator biases can be a major source of error in the analysis of large samples, as those that are becoming increasingly common in the 'era of big data'.
Ivanovska, Tatyana; Bülow, Robin; Biffar, Reiner; Cardini, Andrea
2018-01-01
Using 3D anatomical landmarks from adult human head MRIs, we assessed the magnitude of inter-operator differences in Procrustes-based geometric morphometric analyses. An in depth analysis of both absolute and relative error was performed in a subsample of individuals with replicated digitization by three different operators. The effect of inter-operator differences was also explored in a large sample of more than 900 individuals. Although absolute error was not unusual for MRI measurements, including bone landmarks, shape was particularly affected by differences among operators, with up to more than 30% of sample variation accounted for by this type of error. The magnitude of the bias was such that it dominated the main pattern of bone and total (all landmarks included) shape variation, largely surpassing the effect of sex differences between hundreds of men and women. In contrast, however, we found higher reproducibility in soft-tissue nasal landmarks, despite relatively larger errors in estimates of nasal size. Our study exemplifies the assessment of measurement error using geometric morphometrics on landmarks from MRIs and stresses the importance of relating it to total sample variance within the specific methodological framework being used. In summary, precise landmarks may not necessarily imply negligible errors, especially in shape data; indeed, size and shape may be differentially impacted by measurement error and different types of landmarks may have relatively larger or smaller errors. Importantly, and consistently with other recent studies using geometric morphometrics on digital images (which, however, were not specific to MRI data), this study showed that inter-operator biases can be a major source of error in the analysis of large samples, as those that are becoming increasingly common in the 'era of big data'. PMID:29787586
Sampling errors for a nadir viewing instrument on the International Space Station
NASA Astrophysics Data System (ADS)
Berger, H. I.; Pincus, R.; Evans, F.; Santek, D.; Ackerman, S.; Ackerman, S.
2001-12-01
In an effort to improve the observational charactarization of ice clouds in the earth's atmosphere, we are developing a sub-millimeter wavelength radiometer which we propose to fly on the International Space Station for two years. Our goal is to accurately measure the ice water path and mass-weighted particle size at the finest possible temporal and spatial resolution. The ISS orbit precesses, sampling through the dirunal cycle every 16 days, but technological constraints limit our instrument to a single pixel viewed near nadir. We discuss sampling errors associated with this instrument/platform configuration. We use as "truth" the ISCCP dataset of pixel-level cloud optical retrievals, which acts as a proxy for ice water path; this dataset is sampled according to the orbital characteristics of the space station, and the statistics computed from the sub-sampled population are compared with those from the full dataset. We explore the tradeoffs in average sampling error as a function of the averaging time and spatial scale, and explore the possibility of resolving the dirunal cycle.
Michael, Claire W; Naik, Kalyani; McVicker, Michael
2013-05-01
We developed a value stream map (VSM) of the Papanicolaou test procedure to identify opportunities to reduce waste and errors, created a new VSM, and implemented a new process emphasizing Lean tools. Preimplementation data revealed the following: (1) processing time (PT) for 1,140 samples averaged 54 hours; (2) 27 accessioning errors were detected on review of 357 random requisitions (7.6%); (3) 5 of the 20,060 tests had labeling errors that had gone undetected in the processing stage. Four were detected later during specimen processing but 1 reached the reporting stage. Postimplementation data were as follows: (1) PT for 1,355 samples averaged 31 hours; (2) 17 accessioning errors were detected on review of 385 random requisitions (4.4%); and (3) no labeling errors were undetected. Our results demonstrate that implementation of Lean methods, such as first-in first-out processes and minimizing batch size by staff actively participating in the improvement process, allows for higher quality, greater patient safety, and improved efficiency.
Sullivan, James S.; Ball, Don G.
1997-01-01
The instantaneous V.sub.co signal on a charging capacitor is sampled and the charge voltage on capacitor C.sub.o is captured just prior to its discharge into the first stage of magnetic modulator. The captured signal is applied to an averaging circuit with a long time constant and to the positive input terminal of a differential amplifier. The averaged V.sub. co signal is split between a gain stage (G=0.975) and a feedback stage that determines the slope of the voltage ramp applied to the high speed comparator. The 97.5% portion of the averaged V.sub.co signal is applied to the negative input of a differential amplifier gain stage (G=10). The differential amplifier produces an error signal by subtracting 97.5% of the averaged V.sub.co signal from the instantaneous value of sampled V.sub.co signal and multiplying the difference by ten. The resulting error signal is applied to the positive input of a high speed comparator. The error signal is then compared to a voltage ramp that is proportional to the averaged V.sub.co values squared divided by the total volt-second product of the magnetic compression circuit.
Sullivan, J.S.; Ball, D.G.
1997-09-09
The instantaneous V{sub co} signal on a charging capacitor is sampled and the charge voltage on capacitor C{sub o} is captured just prior to its discharge into the first stage of magnetic modulator. The captured signal is applied to an averaging circuit with a long time constant and to the positive input terminal of a differential amplifier. The averaged V{sub co} signal is split between a gain stage (G = 0.975) and a feedback stage that determines the slope of the voltage ramp applied to the high speed comparator. The 97.5% portion of the averaged V{sub co} signal is applied to the negative input of a differential amplifier gain stage (G = 10). The differential amplifier produces an error signal by subtracting 97.5% of the averaged V{sub co} signal from the instantaneous value of sampled V{sub co} signal and multiplying the difference by ten. The resulting error signal is applied to the positive input of a high speed comparator. The error signal is then compared to a voltage ramp that is proportional to the averaged V{sub co} values squared divided by the total volt-second product of the magnetic compression circuit. 11 figs.
García-González, Miguel A; Fernández-Chimeno, Mireya; Ramos-Castro, Juan
2009-02-01
An analysis of the errors due to the finite resolution of RR time series in the estimation of the approximate entropy (ApEn) is described. The quantification errors in the discrete RR time series produce considerable errors in the ApEn estimation (bias and variance) when the signal variability or the sampling frequency is low. Similar errors can be found in indices related to the quantification of recurrence plots. An easy way to calculate a figure of merit [the signal to resolution of the neighborhood ratio (SRN)] is proposed in order to predict when the bias in the indices could be high. When SRN is close to an integer value n, the bias is higher than when near n - 1/2 or n + 1/2. Moreover, if SRN is close to an integer value, the lower this value, the greater the bias is.
Ji, Xiaohong; Liu, Peng; Sun, Zhenqi; Su, Xiaohui; Wang, Wei; Gao, Yanhui; Sun, Dianjun
2016-01-01
Objective To determine the effect of statistical correction for intra-individual variation on estimated urinary iodine concentration (UIC) by sampling on 3 consecutive days in four seasons in children. Setting School-aged children from urban and rural primary schools in Harbin, Heilongjiang, China. Participants 748 and 640 children aged 8–11 years were recruited from urban and rural schools, respectively, in Harbin. Primary and secondary outcome measures The spot urine samples were collected once a day for 3 consecutive days in each season over 1 year. The UIC of the first day was corrected by two statistical correction methods: the average correction method (average of days 1, 2; average of days 1, 2 and 3) and the variance correction method (UIC of day 1 corrected by two replicates and by three replicates). The variance correction method determined the SD between subjects (Sb) and within subjects (Sw), and calculated the correction coefficient (Fi), Fi=Sb/(Sb+Sw/di), where di was the number of observations. The UIC of day 1 was then corrected using the following equation: Results The variance correction methods showed the overall Fi was 0.742 for 2 days’ correction and 0.829 for 3 days’ correction; the values for the seasons spring, summer, autumn and winter were 0.730, 0.684, 0.706 and 0.703 for 2 days’ correction and 0.809, 0.742, 0.796 and 0.804 for 3 days’ correction, respectively. After removal of the individual effect, the correlation coefficient between consecutive days was 0.224, and between non-consecutive days 0.050. Conclusions The variance correction method is effective for correcting intra-individual variation in estimated UIC following sampling on 3 consecutive days in four seasons in children. The method varies little between ages, sexes and urban or rural setting, but does vary between seasons. PMID:26920442
Ciaccheri, Leonardo; Tuccio, Lorenza; Mencaglia, Andrea A; Sikorska-Zimny, Kalina; Hallmann, Ewelina; Kowalski, Artur; G Mignani, Anna; Kaniszewski, Stanislaw; Agati, Giovanni
2018-05-09
Reflectance spectroscopy represents a useful tool for the nondestructive assessment of tomato lycopene, even in the field. For this reason, a compact, low-cost, light emitting diode-based sensor has been developed to measure reflectance in the 400-750 nm spectral range. It was calibrated against wet chemistry and evaluated by partial least squares (PLS) regression analyses. The lycopene prediction models were defined for two open-field cultivated red-tomato varieties: the processing oblong tomatoes of the cv. Calista (average weight: 76 g) and the fresh-consumption round tomatoes of the cv. Volna (average weight: 130 g), over a period of two consecutive years. The lycopene prediction models were dependent on both cultivar and season. The lycopene root mean square error of prediction produced by the 2014 single-cultivar calibrations validated on the 2015 samples was large (33 mg kg -1 ) in the Calista tomatoes and acceptable (9.5 mg kg -1 ) in the Volna tomatoes. A more general bicultivar and biyear model could still explain almost 80% of the predicted lycopene variance, with a relative error in red tomatoes of less than 20%. In 2016, the in-field applications of the multiseasonal prediction models, built with the 2014 and 2015 data, showed significant ( P < 0.001) differences in the average lycopene estimated in the crop on two sampling dates that were 20 days apart: on August 19 and September 7, 2016, the lycopene was 98.9 ± 9.3 and 92.2 ± 10.8 mg kg -1 FW for cv. Calista and 54.6 ± 13.2 and 60.8 ± 6.8 mg kg -1 FW for cv. Volna. The sensor was also able to monitor the temporal evolution of lycopene accumulation on the very same fruits attached to the plants. These results indicated that a simple, compact reflectance device and PLS analysis could provide adequately precise and robust (through-seasons) models for the nondestructive assessment of lycopene in whole tomatoes. This technique could guarantee tomatoes with the highest nutraceutical value from the production, during storage and distribution, and finally to consumers.
Tracked ultrasound calibration studies with a phantom made of LEGO bricks
NASA Astrophysics Data System (ADS)
Soehl, Marie; Walsh, Ryan; Rankin, Adam; Lasso, Andras; Fichtinger, Gabor
2014-03-01
In this study, spatial calibration of tracked ultrasound was compared by using a calibration phantom made of LEGO® bricks and two 3-D printed N-wire phantoms. METHODS: The accuracy and variance of calibrations were compared under a variety of operating conditions. Twenty trials were performed using an electromagnetic tracking device with a linear probe and three trials were performed using varied probes, varied tracking devices and the three aforementioned phantoms. The accuracy and variance of spatial calibrations found through the standard deviation and error of the 3-D image reprojection were used to compare the calibrations produced from the phantoms. RESULTS: This study found no significant difference between the measured variables of the calibrations. The average standard deviation of multiple 3-D image reprojections with the highest performing printed phantom and those from the phantom made of LEGO® bricks differed by 0.05 mm and the error of the reprojections differed by 0.13 mm. CONCLUSION: Given that the phantom made of LEGO® bricks is significantly less expensive, more readily available, and more easily modified than precision-machined N-wire phantoms, it prompts to be a viable calibration tool especially for quick laboratory research and proof of concept implementations of tracked ultrasound navigation.
Theoretical and simulated performance for a novel frequency estimation technique
NASA Technical Reports Server (NTRS)
Crozier, Stewart N.
1993-01-01
A low complexity, open-loop, discrete-time, delay-multiply-average (DMA) technique for estimating the frequency offset for digitally modulated MPSK signals is investigated. A nonlinearity is used to remove the MPSK modulation and generate the carrier component to be extracted. Theoretical and simulated performance results are presented and compared to the Cramer-Rao lower bound (CRLB) for the variance of the frequency estimation error. For all signal-to-noise ratios (SNR's) above threshold, it is shown that the CRLB can essentially be achieved with linear complexity.
Zheng, Li; Silliman, Stephen E.
2000-01-01
A modification of previously published solutions regarding the spatial variation of hydraulic heads is discussed whereby the semivariogram of increments of head residuals (termed head residual increments HRIs) are related to the variance and integral scale of the transmissivity field. A first‐order solution is developed for the case of a transmissivity field which is isotropic and whose second‐order behavior can be characterized by an exponential covariance structure. The estimates of the variance σY2 and the integral scale λ of the log transmissivity field are then obtained via fitting a theoretical semivariogram for the HRI to its sample semivariogram. This approach is applied to head data sampled from a series of two‐dimensional, simulated aquifers with isotropic, exponential covariance structures and varying degrees of heterogeneity (σY2 = 0.25, 0.5, 1.0, 2.0, and 5.0). The results show that this method provided reliable estimates for both λ and σY2 in aquifers with the value of σY2 up to 2.0, but the errors in those estimates were higher for σY2 equal to 5.0. It is also demonstrated through numerical experiments and theoretical arguments that the head residual increments will provide a sample semivariogram with a lower variance than will the use of the head residuals without calculation of increments.
Dependability of Data Derived from Time Sampling Methods with Multiple Observation Targets
ERIC Educational Resources Information Center
Johnson, Austin H.; Chafouleas, Sandra M.; Briesch, Amy M.
2017-01-01
In this study, generalizability theory was used to examine the extent to which (a) time-sampling methodology, (b) number of simultaneous behavior targets, and (c) individual raters influenced variance in ratings of academic engagement for an elementary-aged student. Ten graduate-student raters, with an average of 7.20 hr of previous training in…
A Late Pleistocene sea level stack
NASA Astrophysics Data System (ADS)
Spratt, Rachel M.; Lisiecki, Lorraine E.
2016-04-01
Late Pleistocene sea level has been reconstructed from ocean sediment core data using a wide variety of proxies and models. However, the accuracy of individual reconstructions is limited by measurement error, local variations in salinity and temperature, and assumptions particular to each technique. Here we present a sea level stack (average) which increases the signal-to-noise ratio of individual reconstructions. Specifically, we perform principal component analysis (PCA) on seven records from 0 to 430 ka and five records from 0 to 798 ka. The first principal component, which we use as the stack, describes ˜ 80 % of the variance in the data and is similar using either five or seven records. After scaling the stack based on Holocene and Last Glacial Maximum (LGM) sea level estimates, the stack agrees to within 5 m with isostatically adjusted coral sea level estimates for Marine Isotope Stages 5e and 11 (125 and 400 ka, respectively). Bootstrapping and random sampling yield mean uncertainty estimates of 9-12 m (1σ) for the scaled stack. Sea level change accounts for about 45 % of the total orbital-band variance in benthic δ18O, compared to a 65 % contribution during the LGM-to-Holocene transition. Additionally, the second and third principal components of our analyses reflect differences between proxy records associated with spatial variations in the δ18O of seawater.
Experimental design, power and sample size for animal reproduction experiments.
Chapman, Phillip L; Seidel, George E
2008-01-01
The present paper concerns statistical issues in the design of animal reproduction experiments, with emphasis on the problems of sample size determination and power calculations. We include examples and non-technical discussions aimed at helping researchers avoid serious errors that may invalidate or seriously impair the validity of conclusions from experiments. Screen shots from interactive power calculation programs and basic SAS power calculation programs are presented to aid in understanding statistical power and computing power in some common experimental situations. Practical issues that are common to most statistical design problems are briefly discussed. These include one-sided hypothesis tests, power level criteria, equality of within-group variances, transformations of response variables to achieve variance equality, optimal specification of treatment group sizes, 'post hoc' power analysis and arguments for the increased use of confidence intervals in place of hypothesis tests.
Organizational safety culture and medical error reporting by Israeli nurses.
Kagan, Ilya; Barnoy, Sivia
2013-09-01
To investigate the association between patient safety culture (PSC) and the incidence and reporting rate of medical errors by Israeli nurses. Self-administered structured questionnaires were distributed to a convenience sample of 247 registered nurses enrolled in training programs at Tel Aviv University (response rate = 91%). The questionnaire's three sections examined the incidence of medication mistakes in clinical practice, the reporting rate for these errors, and the participants' views and perceptions of the safety culture in their workplace at three levels (organizational, departmental, and individual performance). Pearson correlation coefficients, t tests, and multiple regression analysis were used to analyze the data. Most nurses encountered medical errors from a daily to a weekly basis. Six percent of the sample never reported their own errors, while half reported their own errors "rarely or sometimes." The level of PSC was positively and significantly correlated with the error reporting rate. PSC, place of birth, error incidence, and not having an academic nursing degree were significant predictors of error reporting, together explaining 28% of variance. This study confirms the influence of an organizational safety climate on readiness to report errors. Senior healthcare executives and managers can make a major impact on safety culture development by creating and promoting a vision and strategy for quality and safety and fostering their employees' motivation to implement improvement programs at the departmental and individual level. A positive, carefully designed organizational safety culture can encourage error reporting by staff and so improve patient safety. © 2013 Sigma Theta Tau International.
Styck, Kara M; Walsh, Shana M
2016-01-01
The purpose of the present investigation was to conduct a meta-analysis of the literature on examiner errors for the Wechsler scales of intelligence. Results indicate that a mean of 99.7% of protocols contained at least 1 examiner error when studies that included a failure to record examinee responses as an error were combined and a mean of 41.2% of protocols contained at least 1 examiner error when studies that ignored errors of omission were combined. Furthermore, graduate student examiners were significantly more likely to make at least 1 error on Wechsler intelligence test protocols than psychologists. However, psychologists made significantly more errors per protocol than graduate student examiners regardless of the inclusion or exclusion of failure to record examinee responses as errors. On average, 73.1% of Full-Scale IQ (FSIQ) scores changed as a result of examiner errors, whereas 15.8%-77.3% of scores on the Verbal Comprehension Index (VCI), Perceptual Reasoning Index (PRI), Working Memory Index (WMI), and Processing Speed Index changed as a result of examiner errors. In addition, results suggest that examiners tend to overestimate FSIQ scores and underestimate VCI scores. However, no strong pattern emerged for the PRI and WMI. It can be concluded that examiner errors occur frequently and impact index and FSIQ scores. Consequently, current estimates for the standard error of measurement of popular IQ tests may not adequately capture the variance due to the examiner. (c) 2016 APA, all rights reserved).
Sources of error in estimating truck traffic from automatic vehicle classification data
DOT National Transportation Integrated Search
1998-10-01
Truck annual average daily traffic estimation errors resulting from sample classification counts are computed in this paper under two scenarios. One scenario investigates an improper factoring procedure that may be used by highway agencies. The study...
Reconstruction of regional mean temperature for East Asia since 1900s and its uncertainties
NASA Astrophysics Data System (ADS)
Hua, W.
2017-12-01
Regional average surface air temperature (SAT) is one of the key variables often used to investigate climate change. Unfortunately, because of the limited observations over East Asia, there were also some gaps in the observation data sampling for regional mean SAT analysis, which was important to estimate past climate change. In this study, the regional average temperature of East Asia since 1900s is calculated by the Empirical Orthogonal Function (EOF)-based optimal interpolation (OA) method with considering the data errors. The results show that our estimate is more precise and robust than the results from simple average, which provides a better way for past climate reconstruction. In addition to the reconstructed regional average SAT anomaly time series, we also estimated uncertainties of reconstruction. The root mean square error (RMSE) results show that the the error decreases with respect to time, and are not sufficiently large to alter the conclusions on the persist warming in East Asia during twenty-first century. Moreover, the test of influence of data error on reconstruction clearly shows the sensitivity of reconstruction to the size of the data error.
NASA Astrophysics Data System (ADS)
Pietrzyk, Mariusz W.; Donovan, Tim; Brennan, Patrick C.; Dix, Alan; Manning, David J.
2011-03-01
Aim: To optimize automated classification of radiological errors during lung nodule detection from chest radiographs (CxR) using a support vector machine (SVM) run on the spatial frequency features extracted from the local background of selected regions. Background: The majority of the unreported pulmonary nodules are visually detected but not recognized; shown by the prolonged dwell time values at false-negative regions. Similarly, overestimated nodule locations are capturing substantial amounts of foveal attention. Spatial frequency properties of selected local backgrounds are correlated with human observer responses either in terms of accuracy in indicating abnormality position or in the precision of visual sampling the medical images. Methods: Seven radiologists participated in the eye tracking experiments conducted under conditions of pulmonary nodule detection from a set of 20 postero-anterior CxR. The most dwelled locations have been identified and subjected to spatial frequency (SF) analysis. The image-based features of selected ROI were extracted with un-decimated Wavelet Packet Transform. An analysis of variance was run to select SF features and a SVM schema was implemented to classify False-Negative and False-Positive from all ROI. Results: A relative high overall accuracy was obtained for each individually developed Wavelet-SVM algorithm, with over 90% average correct ratio for errors recognition from all prolonged dwell locations. Conclusion: The preliminary results show that combined eye-tracking and image-based features can be used for automated detection of radiological error with SVM. The work is still in progress and not all analytical procedures have been completed, which might have an effect on the specificity of the algorithm.
Bischel, Alexander; Stratis, Andreas; Kakar, Apoorv; Bosmans, Hilde; Jacobs, Reinhilde; Gassner, Eva-Maria; Puelacher, Wolfgang; Pauwels, Ruben
2016-01-01
Objective: The aim of this study was to evaluate whether application of ultralow dose protocols and iterative reconstruction technology (IRT) influence quantitative Hounsfield units (HUs) and contrast-to-noise ratio (CNR) in dentomaxillofacial CT imaging. Methods: A phantom with inserts of five types of materials was scanned using protocols for (a) a clinical reference for navigated surgery (CT dose index volume 36.58 mGy), (b) low-dose sinus imaging (18.28 mGy) and (c) four ultralow dose imaging (4.14, 2.63, 0.99 and 0.53 mGy). All images were reconstructed using: (i) filtered back projection (FBP); (ii) IRT: adaptive statistical iterative reconstruction-50 (ASIR-50), ASIR-100 and model-based iterative reconstruction (MBIR); and (iii) standard (std) and bone kernel. Mean HU, CNR and average HU error after recalibration were determined. Each combination of protocols was compared using Friedman analysis of variance, followed by Dunn's multiple comparison test. Results: Pearson's sample correlation coefficients were all >0.99. Ultralow dose protocols using FBP showed errors of up to 273 HU. Std kernels had less HU variability than bone kernels. MBIR reduced the error value for the lowest dose protocol to 138 HU and retained the highest relative CNR. ASIR could not demonstrate significant advantages over FBP. Conclusions: Considering a potential dose reduction as low as 1.5% of a std protocol, ultralow dose protocols and IRT should be further tested for clinical dentomaxillofacial CT imaging. Advances in knowledge: HU as a surrogate for bone density may vary significantly in CT ultralow dose imaging. However, use of std kernels and MBIR technology reduce HU error values and may retain the highest CNR. PMID:26859336
NASA Technical Reports Server (NTRS)
North, G. R.; Bell, T. L.; Cahalan, R. F.; Moeng, F. J.
1982-01-01
Geometric characteristics of the spherical earth are shown to be responsible for the increase of variance with latitude of zonally averaged meteorological statistics. An analytic model is constructed to display the effect of a spherical geometry on zonal averages, employing a sphere labeled with radial unit vectors in a real, stochastic field expanded in complex spherical harmonics. The variance of a zonally averaged field is found to be expressible in terms of the spectrum of the vector field of the spherical harmonics. A maximum variance is then located at the poles, and the ratio of the variance to the zonally averaged grid-point variance, weighted by the cosine of the latitude, yields the zonal correlation typical of the latitude. An example is provided for the 500 mb level in the Northern Hemisphere compared to 15 years of data. Variance is determined to increase north of 60 deg latitude.
Wang, Chunhao; Yin, Fang-Fang; Kirkpatrick, John P; Chang, Zheng
2017-08-01
To investigate the feasibility of using undersampled k-space data and an iterative image reconstruction method with total generalized variation penalty in the quantitative pharmacokinetic analysis for clinical brain dynamic contrast-enhanced magnetic resonance imaging. Eight brain dynamic contrast-enhanced magnetic resonance imaging scans were retrospectively studied. Two k-space sparse sampling strategies were designed to achieve a simulated image acquisition acceleration factor of 4. They are (1) a golden ratio-optimized 32-ray radial sampling profile and (2) a Cartesian-based random sampling profile with spatiotemporal-regularized sampling density constraints. The undersampled data were reconstructed to yield images using the investigated reconstruction technique. In quantitative pharmacokinetic analysis on a voxel-by-voxel basis, the rate constant K trans in the extended Tofts model and blood flow F B and blood volume V B from the 2-compartment exchange model were analyzed. Finally, the quantitative pharmacokinetic parameters calculated from the undersampled data were compared with the corresponding calculated values from the fully sampled data. To quantify each parameter's accuracy calculated using the undersampled data, error in volume mean, total relative error, and cross-correlation were calculated. The pharmacokinetic parameter maps generated from the undersampled data appeared comparable to the ones generated from the original full sampling data. Within the region of interest, most derived error in volume mean values in the region of interest was about 5% or lower, and the average error in volume mean of all parameter maps generated through either sampling strategy was about 3.54%. The average total relative error value of all parameter maps in region of interest was about 0.115, and the average cross-correlation of all parameter maps in region of interest was about 0.962. All investigated pharmacokinetic parameters had no significant differences between the result from original data and the reduced sampling data. With sparsely sampled k-space data in simulation of accelerated acquisition by a factor of 4, the investigated dynamic contrast-enhanced magnetic resonance imaging pharmacokinetic parameters can accurately estimate the total generalized variation-based iterative image reconstruction method for reliable clinical application.
Harris, Alexandre M.; DeGiorgio, Michael
2016-01-01
Gene diversity, or expected heterozygosity (H), is a common statistic for assessing genetic variation within populations. Estimation of this statistic decreases in accuracy and precision when individuals are related or inbred, due to increased dependence among allele copies in the sample. The original unbiased estimator of expected heterozygosity underestimates true population diversity in samples containing relatives, as it only accounts for sample size. More recently, a general unbiased estimator of expected heterozygosity was developed that explicitly accounts for related and inbred individuals in samples. Though unbiased, this estimator’s variance is greater than that of the original estimator. To address this issue, we introduce a general unbiased estimator of gene diversity for samples containing related or inbred individuals, which employs the best linear unbiased estimator of allele frequencies, rather than the commonly used sample proportion. We examine the properties of this estimator, H∼BLUE, relative to alternative estimators using simulations and theoretical predictions, and show that it predominantly has the smallest mean squared error relative to others. Further, we empirically assess the performance of H∼BLUE on a global human microsatellite dataset of 5795 individuals, from 267 populations, genotyped at 645 loci. Additionally, we show that the improved variance of H∼BLUE leads to improved estimates of the population differentiation statistic, FST, which employs measures of gene diversity within its calculation. Finally, we provide an R script, BestHet, to compute this estimator from genomic and pedigree data. PMID:28040781
Sampling errors in the measurement of rain and hail parameters
NASA Technical Reports Server (NTRS)
Gertzman, H. S.; Atlas, D.
1977-01-01
Attention is given to a general derivation of the fractional standard deviation (FSD) of any integrated property X such that X(D) = cD to the n. This work extends that of Joss and Waldvogel (1969). The equation is applicable to measuring integrated properties of cloud, rain or hail populations (such as water content, precipitation rate, kinetic energy, or radar reflectivity) which are subject to statistical sampling errors due to the Poisson distributed fluctuations of particles sampled in each particle size interval and the weighted sum of the associated variances in proportion to their contribution to the integral parameter to be measured. Universal curves are presented which are applicable to the exponential size distribution permitting FSD estimation of any parameters from n = 0 to n = 6. The equations and curves also permit corrections for finite upper limits in the size spectrum and a realistic fall speed law.
Reliability of a Longitudinal Sequence of Scale Ratings
ERIC Educational Resources Information Center
Laenen, Annouschka; Alonso, Ariel; Molenberghs, Geert; Vangeneugden, Tony
2009-01-01
Reliability captures the influence of error on a measurement and, in the classical setting, is defined as one minus the ratio of the error variance to the total variance. Laenen, Alonso, and Molenberghs ("Psychometrika" 73:443-448, 2007) proposed an axiomatic definition of reliability and introduced the R[subscript T] coefficient, a measure of…
Impact of Measurement Error on Statistical Power: Review of an Old Paradox.
ERIC Educational Resources Information Center
Williams, Richard H.; And Others
1995-01-01
The paradox that a Student t-test based on pretest-posttest differences can attain its greatest power when the difference score reliability is zero was explained by demonstrating that power is not a mathematical function of reliability unless either true score variance or error score variance is constant. (SLD)
High-Dimensional Heteroscedastic Regression with an Application to eQTL Data Analysis
Daye, Z. John; Chen, Jinbo; Li, Hongzhe
2011-01-01
Summary We consider the problem of high-dimensional regression under non-constant error variances. Despite being a common phenomenon in biological applications, heteroscedasticity has, so far, been largely ignored in high-dimensional analysis of genomic data sets. We propose a new methodology that allows non-constant error variances for high-dimensional estimation and model selection. Our method incorporates heteroscedasticity by simultaneously modeling both the mean and variance components via a novel doubly regularized approach. Extensive Monte Carlo simulations indicate that our proposed procedure can result in better estimation and variable selection than existing methods when heteroscedasticity arises from the presence of predictors explaining error variances and outliers. Further, we demonstrate the presence of heteroscedasticity in and apply our method to an expression quantitative trait loci (eQTLs) study of 112 yeast segregants. The new procedure can automatically account for heteroscedasticity in identifying the eQTLs that are associated with gene expression variations and lead to smaller prediction errors. These results demonstrate the importance of considering heteroscedasticity in eQTL data analysis. PMID:22547833
Accounting for measurement error: a critical but often overlooked process.
Harris, Edward F; Smith, Richard N
2009-12-01
Due to instrument imprecision and human inconsistencies, measurements are not free of error. Technical error of measurement (TEM) is the variability encountered between dimensions when the same specimens are measured at multiple sessions. A goal of a data collection regimen is to minimise TEM. The few studies that actually quantify TEM, regardless of discipline, report that it is substantial and can affect results and inferences. This paper reviews some statistical approaches for identifying and controlling TEM. Statistically, TEM is part of the residual ('unexplained') variance in a statistical test, so accounting for TEM, which requires repeated measurements, enhances the chances of finding a statistically significant difference if one exists. The aim of this paper was to review and discuss common statistical designs relating to types of error and statistical approaches to error accountability. This paper addresses issues of landmark location, validity, technical and systematic error, analysis of variance, scaled measures and correlation coefficients in order to guide the reader towards correct identification of true experimental differences. Researchers commonly infer characteristics about populations from comparatively restricted study samples. Most inferences are statistical and, aside from concerns about adequate accounting for known sources of variation with the research design, an important source of variability is measurement error. Variability in locating landmarks that define variables is obvious in odontometrics, cephalometrics and anthropometry, but the same concerns about measurement accuracy and precision extend to all disciplines. With increasing accessibility to computer-assisted methods of data collection, the ease of incorporating repeated measures into statistical designs has improved. Accounting for this technical source of variation increases the chance of finding biologically true differences when they exist.
Statistical power for detecting trends with applications to seabird monitoring
Hatch, Shyla A.
2003-01-01
Power analysis is helpful in defining goals for ecological monitoring and evaluating the performance of ongoing efforts. I examined detection standards proposed for population monitoring of seabirds using two programs (MONITOR and TRENDS) specially designed for power analysis of trend data. Neither program models within- and among-years components of variance explicitly and independently, thus an error term that incorporates both components is an essential input. Residual variation in seabird counts consisted of day-to-day variation within years and unexplained variation among years in approximately equal parts. The appropriate measure of error for power analysis is the standard error of estimation (S.E.est) from a regression of annual means against year. Replicate counts within years are helpful in minimizing S.E.est but should not be treated as independent samples for estimating power to detect trends. Other issues include a choice of assumptions about variance structure and selection of an exponential or linear model of population change. Seabird count data are characterized by strong correlations between S.D. and mean, thus a constant CV model is appropriate for power calculations. Time series were fit about equally well with exponential or linear models, but log transformation ensures equal variances over time, a basic assumption of regression analysis. Using sample data from seabird monitoring in Alaska, I computed the number of years required (with annual censusing) to detect trends of -1.4% per year (50% decline in 50 years) and -2.7% per year (50% decline in 25 years). At ??=0.05 and a desired power of 0.9, estimated study intervals ranged from 11 to 69 years depending on species, trend, software, and study design. Power to detect a negative trend of 6.7% per year (50% decline in 10 years) is suggested as an alternative standard for seabird monitoring that achieves a reasonable match between statistical and biological significance.
NASA Astrophysics Data System (ADS)
Kinnard, Lisa M.; Gavrielides, Marios A.; Myers, Kyle J.; Zeng, Rongping; Peregoy, Jennifer; Pritchard, William; Karanian, John W.; Petrick, Nicholas
2008-03-01
High-resolution CT, three-dimensional (3D) methods for nodule volumetry have been introduced, with the hope that such methods will be more accurate and consistent than currently used planar measures of size. However, the error associated with volume estimation methods still needs to be quantified. Volume estimation error is multi-faceted in the sense that it is impacted by characteristics of the patient, the software tool and the CT system. The overall goal of this research is to quantify the various sources of measurement error and, when possible, minimize their effects. In the current study, we estimated nodule volume from ten repeat scans of an anthropomorphic phantom containing two synthetic spherical lung nodules (diameters: 5 and 10 mm; density: -630 HU), using a 16-slice Philips CT with 20, 50, 100 and 200 mAs exposures and 0.8 and 3.0 mm slice thicknesses. True volume was estimated from an average of diameter measurements, made using digital calipers. We report variance and bias results for volume measurements as a function of slice thickness, nodule diameter, and X-ray exposure.
Grogger, P; Sacher, C; Weber, S; Millesi, G; Seemann, R
2018-04-10
Deviations in measuring dentofacial components in a lateral X-ray represent a major hurdle in the subsequent treatment of dysgnathic patients. In a retrospective study, we investigated the most prevalent source of error in the following commonly used cephalometric measurements: the angles Sella-Nasion-Point A (SNA), Sella-Nasion-Point B (SNB) and Point A-Nasion-Point B (ANB); the Wits appraisal; the anteroposterior dysplasia indicator (APDI); and the overbite depth indicator (ODI). Preoperative lateral radiographic images of patients with dentofacial deformities were collected and the landmarks digitally traced by three independent raters. Cephalometric analysis was automatically performed based on 1116 tracings. Error analysis identified the x-coordinate of Point A as the prevalent source of error in all investigated measurements, except SNB, in which it is not incorporated. In SNB, the y-coordinate of Nasion predominated error variance. SNB showed lowest inter-rater variation. In addition, our observations confirmed previous studies showing that landmark identification variance follows characteristic error envelopes in the highest number of tracings analysed up to now. Variance orthogonal to defining planes was of relevance, while variance parallel to planes was not. Taking these findings into account, orthognathic surgeons as well as orthodontists would be able to perform cephalometry more accurately and accomplish better therapeutic results. Copyright © 2018 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Analysis of Realized Volatility for Nikkei Stock Average on the Tokyo Stock Exchange
NASA Astrophysics Data System (ADS)
Takaishi, Tetsuya; Watanabe, Toshiaki
2016-04-01
We calculate realized volatility of the Nikkei Stock Average (Nikkei225) Index on the Tokyo Stock Exchange and investigate the return dynamics. To avoid the bias on the realized volatility from the non-trading hours issue we calculate realized volatility separately in the two trading sessions, i.e. morning and afternoon, of the Tokyo Stock Exchange and find that the microstructure noise decreases the realized volatility at small sampling frequency. Using realized volatility as a proxy of the integrated volatility we standardize returns in the morning and afternoon sessions and investigate the normality of the standardized returns by calculating variance, kurtosis and 6th moment. We find that variance, kurtosis and 6th moment are consistent with those of the standard normal distribution, which indicates that the return dynamics of the Nikkei Stock Average are well described by a Gaussian random process with time-varying volatility.
Hinton-Bayre, Anton D
2011-02-01
There is an ongoing debate over the preferred method(s) for determining the reliable change (RC) in individual scores over time. In the present paper, specificity comparisons of several classic and contemporary RC models were made using a real data set. This included a more detailed review of a new RC model recently proposed in this journal, that used the within-subjects standard deviation (WSD) as the error term. It was suggested that the RC(WSD) was more sensitive to change and theoretically superior. The current paper demonstrated that even in the presence of mean practice effects, false-positive rates were comparable across models when reliability was good and initial and retest variances were equivalent. However, when variances differed, discrepancies in classification across models became evident. Notably, the RC using the WSD provided unacceptably high false-positive rates in this setting. It was considered that the WSD was never intended for measuring change in this manner. The WSD actually combines systematic and error variance. The systematic variance comes from measurable between-treatment differences, commonly referred to as practice effect. It was further demonstrated that removal of the systematic variance and appropriate modification of the residual error term for the purpose of testing individual change yielded an error term already published and criticized in the literature. A consensus on the RC approach is needed. To that end, further comparison of models under varied conditions is encouraged.
NASA Astrophysics Data System (ADS)
Bindschadler, Michael; Modgil, Dimple; Branch, Kelley R.; La Riviere, Patrick J.; Alessio, Adam M.
2014-04-01
Myocardial blood flow (MBF) can be estimated from dynamic contrast enhanced (DCE) cardiac CT acquisitions, leading to quantitative assessment of regional perfusion. The need for low radiation dose and the lack of consensus on MBF estimation methods motivates this study to refine the selection of acquisition protocols and models for CT-derived MBF. DCE cardiac CT acquisitions were simulated for a range of flow states (MBF = 0.5, 1, 2, 3 ml (min g)-1, cardiac output = 3, 5, 8 L min-1). Patient kinetics were generated by a mathematical model of iodine exchange incorporating numerous physiological features including heterogenenous microvascular flow, permeability and capillary contrast gradients. CT acquisitions were simulated for multiple realizations of realistic x-ray flux levels. CT acquisitions that reduce radiation exposure were implemented by varying both temporal sampling (1, 2, and 3 s sampling intervals) and tube currents (140, 70, and 25 mAs). For all acquisitions, we compared three quantitative MBF estimation methods (two-compartment model, an axially-distributed model, and the adiabatic approximation to the tissue homogeneous model) and a qualitative slope-based method. In total, over 11 000 time attenuation curves were used to evaluate MBF estimation in multiple patient and imaging scenarios. After iodine-based beam hardening correction, the slope method consistently underestimated flow by on average 47.5% and the quantitative models provided estimates with less than 6.5% average bias and increasing variance with increasing dose reductions. The three quantitative models performed equally well, offering estimates with essentially identical root mean squared error (RMSE) for matched acquisitions. MBF estimates using the qualitative slope method were inferior in terms of bias and RMSE compared to the quantitative methods. MBF estimate error was equal at matched dose reductions for all quantitative methods and range of techniques evaluated. This suggests that there is no particular advantage between quantitative estimation methods nor to performing dose reduction via tube current reduction compared to temporal sampling reduction. These data are important for optimizing implementation of cardiac dynamic CT in clinical practice and in prospective CT MBF trials.
Performance of the all-digital data-transition tracking loop in the advanced receiver
NASA Astrophysics Data System (ADS)
Cheng, U.; Hinedi, S.
1989-11-01
The performance of the all-digital data-transition tracking loop (DTTL) with coherent or noncoherent sampling is described. The effects of few samples per symbol and of noncommensurate sampling rates and symbol rates are addressed and analyzed. Their impacts on the loop phase-error variance and the mean time to lose lock (MTLL) are quantified through computer simulations. The analysis and preliminary simulations indicate that with three to four samples per symbol, the DTTL can track with negligible jitter because of the presence of earth Doppler rate. Furthermore, the MTLL is also expected to be large engough to maintain lock over a Deep Space Network track.
Simplified Estimation and Testing in Unbalanced Repeated Measures Designs.
Spiess, Martin; Jordan, Pascal; Wendt, Mike
2018-05-07
In this paper we propose a simple estimator for unbalanced repeated measures design models where each unit is observed at least once in each cell of the experimental design. The estimator does not require a model of the error covariance structure. Thus, circularity of the error covariance matrix and estimation of correlation parameters and variances are not necessary. Together with a weak assumption about the reason for the varying number of observations, the proposed estimator and its variance estimator are unbiased. As an alternative to confidence intervals based on the normality assumption, a bias-corrected and accelerated bootstrap technique is considered. We also propose the naive percentile bootstrap for Wald-type tests where the standard Wald test may break down when the number of observations is small relative to the number of parameters to be estimated. In a simulation study we illustrate the properties of the estimator and the bootstrap techniques to calculate confidence intervals and conduct hypothesis tests in small and large samples under normality and non-normality of the errors. The results imply that the simple estimator is only slightly less efficient than an estimator that correctly assumes a block structure of the error correlation matrix, a special case of which is an equi-correlation matrix. Application of the estimator and the bootstrap technique is illustrated using data from a task switch experiment based on an experimental within design with 32 cells and 33 participants.
Predicting negative drinking consequences: examining descriptive norm perception.
Benton, Stephen L; Downey, Ronald G; Glider, Peggy S; Benton, Sherry A; Shin, Kanghyun; Newton, Douglas W; Arck, William; Price, Amy
2006-05-01
This study explored how much variance in college student negative drinking consequences is explained by descriptive norm perception, beyond that accounted for by student gender and self-reported alcohol use. A derivation sample (N=7565; 54% women) and a replication sample (N=8924; 55.5% women) of undergraduate students completed the Campus Alcohol Survey in classroom settings. Hierarchical regression analyses revealed that student gender and average number of drinks when "partying" were significantly related to harmful consequences resulting from drinking. Men reported more consequences than did women, and drinking amounts were positively correlated with consequences. However, descriptive norm perception did not explain any additional variance beyond that attributed to gender and alcohol use. Furthermore, there was no significant three-way interaction among student gender, alcohol use, and descriptive norm perception. Norm perception contributed no significant variance in explaining harmful consequences beyond that explained by college student gender and alcohol use.
Hevesi, Joseph A.; Flint, Alan L.; Istok, Jonathan D.
1992-01-01
Values of average annual precipitation (AAP) may be important for hydrologic characterization of a potential high-level nuclear-waste repository site at Yucca Mountain, Nevada. Reliable measurements of AAP are sparse in the vicinity of Yucca Mountain, and estimates of AAP were needed for an isohyetal mapping over a 2600-square-mile watershed containing Yucca Mountain. Estimates were obtained with a multivariate geostatistical model developed using AAP and elevation data from a network of 42 precipitation stations in southern Nevada and southeastern California. An additional 1531 elevations were obtained to improve estimation accuracy. Isohyets representing estimates obtained using univariate geostatistics (kriging) defined a smooth and continuous surface. Isohyets representing estimates obtained using multivariate geostatistics (cokriging) defined an irregular surface that more accurately represented expected local orographic influences on AAP. Cokriging results included a maximum estimate within the study area of 335 mm at an elevation of 7400 ft, an average estimate of 157 mm for the study area, and an average estimate of 172 mm at eight locations in the vicinity of the potential repository site. Kriging estimates tended to be lower in comparison because the increased AAP expected for remote mountainous topography was not adequately represented by the available sample. Regression results between cokriging estimates and elevation were similar to regression results between measured AAP and elevation. The position of the cokriging 250-mm isohyet relative to the boundaries of pinyon pine and juniper woodlands provided indirect evidence of improved estimation accuracy because the cokriging result agreed well with investigations by others concerning the relationship between elevation, vegetation, and climate in the Great Basin. Calculated estimation variances were also mapped and compared to evaluate improvements in estimation accuracy. Cokriging estimation variances were reduced by an average of 54% relative to kriging variances within the study area. Cokriging reduced estimation variances at the potential repository site by 55% relative to kriging. The usefulness of an existing network of stations for measuring AAP within the study area was evaluated using cokriging variances, and twenty additional stations were located for the purpose of improving the accuracy of future isohyetal mappings. Using the expanded network of stations, the maximum cokriging estimation variance within the study area was reduced by 78% relative to the existing network, and the average estimation variance was reduced by 52%.
Aperture averaging and BER for Gaussian beam in underwater oceanic turbulence
NASA Astrophysics Data System (ADS)
Gökçe, Muhsin Caner; Baykal, Yahya
2018-03-01
In an underwater wireless optical communication (UWOC) link, power fluctuations over finite-sized collecting lens are investigated for a horizontally propagating Gaussian beam wave. The power scintillation index, also known as the irradiance flux variance, for the received irradiance is evaluated in weak oceanic turbulence by using the Rytov method. This lets us further quantify the associated performance indicators, namely, the aperture averaging factor and the average bit-error rate (
Optimal sensor fusion for land vehicle navigation
DOE Office of Scientific and Technical Information (OSTI.GOV)
Morrow, J.D.
1990-10-01
Position location is a fundamental requirement in autonomous mobile robots which record and subsequently follow x,y paths. The Dept. of Energy, Office of Safeguards and Security, Robotic Security Vehicle (RSV) program involves the development of an autonomous mobile robot for patrolling a structured exterior environment. A straight-forward method for autonomous path-following has been adopted and requires digitizing'' the desired road network by storing x,y coordinates every 2m along the roads. The position location system used to define the locations consists of a radio beacon system which triangulates position off two known transponders, and dead reckoning with compass and odometer. Thismore » paper addresses the problem of combining these two measurements to arrive at a best estimate of position. Two algorithms are proposed: the optimal'' algorithm treats the measurements as random variables and minimizes the estimate variance, while the average error'' algorithm considers the bias in dead reckoning and attempts to guarantee an average error. Data collected on the algorithms indicate that both work well in practice. 2 refs., 7 figs.« less
Evaluation of Eight Methods for Aligning Orientation of Two Coordinate Systems.
Mecheri, Hakim; Robert-Lachaine, Xavier; Larue, Christian; Plamondon, André
2016-08-01
The aim of this study was to evaluate eight methods for aligning the orientation of two different local coordinate systems. Alignment is very important when combining two different systems of motion analysis. Two of the methods were developed specifically for biomechanical studies, and because there have been at least three decades of algorithm development in robotics, it was decided to include six methods from this field. To compare these methods, an Xsens sensor and two Optotrak clusters were attached to a Plexiglas plate. The first optical marker cluster was fixed on the sensor and 20 trials were recorded. The error of alignment was calculated for each trial, and the mean, the standard deviation, and the maximum values of this error over all trials were reported. One-way repeated measures analysis of variance revealed that the alignment error differed significantly across the eight methods. Post-hoc tests showed that the alignment error from the methods based on angular velocities was significantly lower than for the other methods. The method using angular velocities performed the best, with an average error of 0.17 ± 0.08 deg. We therefore recommend this method, which is easy to perform and provides accurate alignment.
Lin, P.-S.; Chiou, B.; Abrahamson, N.; Walling, M.; Lee, C.-T.; Cheng, C.-T.
2011-01-01
In this study, we quantify the reduction in the standard deviation for empirical ground-motion prediction models by removing ergodic assumption.We partition the modeling error (residual) into five components, three of which represent the repeatable source-location-specific, site-specific, and path-specific deviations from the population mean. A variance estimation procedure of these error components is developed for use with a set of recordings from earthquakes not heavily clustered in space.With most source locations and propagation paths sampled only once, we opt to exploit the spatial correlation of residuals to estimate the variances associated with the path-specific and the source-location-specific deviations. The estimation procedure is applied to ground-motion amplitudes from 64 shallow earthquakes in Taiwan recorded at 285 sites with at least 10 recordings per site. The estimated variance components are used to quantify the reduction in aleatory variability that can be used in hazard analysis for a single site and for a single path. For peak ground acceleration and spectral accelerations at periods of 0.1, 0.3, 0.5, 1.0, and 3.0 s, we find that the singlesite standard deviations are 9%-14% smaller than the total standard deviation, whereas the single-path standard deviations are 39%-47% smaller.
Heat balances of the surface mixed layer in the equatorial Atlantic and Indian Ocean during FGGE
NASA Technical Reports Server (NTRS)
Molinari, R. L.
1985-01-01
Surface meteorological and surface and subsurface oceanographic data collected during FGGE in the equatorial Atlantic and Indian Oceans are used to estimate the terms in a heat balance relation for the mixed layer. The first balance tested is between changes in mixed layer temperature (MLT) and surface energy fluxes. Away from regions of low variance in MLT time series and equatorial and coastal upwelling, surface fluxes can account for 75 percent of the variance in the observed time series. Differences between observed and estimated MLTs indicate that on the average, maximum errors in surface flux are of the order of 20 to 30 W/sq m. In the Atlantic, the addition of zonal advection does not significantly improve the estimates. However in regions of equatorial upwelling, the eastern Atlantic vertical mixing and meridional advection can play an important role in the evolution of MLTs.
The performance of the standard rate turn (SRT) by student naval helicopter pilots.
Chapman, F; Temme, L A; Still, D L
2001-04-01
During flight training, student naval helicopter pilots learn the use of flight instruments through a prescribed series of simulator training events. The training simulator is a 6-degrees-of-freedom, motion-based, high-fidelity instrument trainer. From the final basic instrument simulator flights of student pilots, we selected for evaluation and analysis their performance of the Standard Rate Turn (SRT), a routine flight maneuver. The performance of the SRT was scored with air speed, altitude and heading average error from target values and standard deviations. These average errors and standard deviations were used in a Multiple Analysis of Variance (MANOVA) to evaluate the effects of three independent variables: 1) direction of turn (left vs. right), 2) degree of turn (180 vs. 360 degrees); and 3) segment of turn (roll-in, first 30 s, last 30 s, and roll-out of turn). Only the main effects of the three independent variables were significant; there were no significant interactions. This result greatly reduces the number of different conditions that should be scored separately for the evaluation of SRT performance. The results also showed that the magnitude of the heading and altitude errors at the beginning of the SRT correlated with the magnitude of the heading and altitude errors throughout the turn. This result suggests that for the turn to be well executed, it is important for it to begin with little error in these two response parameters. The observations reported here should be considered when establishing SRT performance norms and comparing student scores. Furthermore, it seems easier for pilots to maintain good performance than to correct poor performance.
RFI in hybrid loops - Simulation and experimental results.
NASA Technical Reports Server (NTRS)
Ziemer, R. E.; Nelson, D. R.; Raghavan, H. R.
1972-01-01
A digital simulation of an imperfect second-order hybrid phase-locked loop (HPLL) operating in radio frequency interference (RFI) is described. Its performance is characterized in terms of phase error variance and phase error probability density function (PDF). Monte-Carlo simulation is used to show that the HPLL can be superior to the conventional phase-locked loops in RFI backgrounds when minimum phase error variance is the goodness criterion. Similar experimentally obtained data are given in support of the simulation data.
A log-sinh transformation for data normalization and variance stabilization
NASA Astrophysics Data System (ADS)
Wang, Q. J.; Shrestha, D. L.; Robertson, D. E.; Pokhrel, P.
2012-05-01
When quantifying model prediction uncertainty, it is statistically convenient to represent model errors that are normally distributed with a constant variance. The Box-Cox transformation is the most widely used technique to normalize data and stabilize variance, but it is not without limitations. In this paper, a log-sinh transformation is derived based on a pattern of errors commonly seen in hydrological model predictions. It is suited to applications where prediction variables are positively skewed and the spread of errors is seen to first increase rapidly, then slowly, and eventually approach a constant as the prediction variable becomes greater. The log-sinh transformation is applied in two case studies, and the results are compared with one- and two-parameter Box-Cox transformations.
An analytic technique for statistically modeling random atomic clock errors in estimation
NASA Technical Reports Server (NTRS)
Fell, P. J.
1981-01-01
Minimum variance estimation requires that the statistics of random observation errors be modeled properly. If measurements are derived through the use of atomic frequency standards, then one source of error affecting the observable is random fluctuation in frequency. This is the case, for example, with range and integrated Doppler measurements from satellites of the Global Positioning and baseline determination for geodynamic applications. An analytic method is presented which approximates the statistics of this random process. The procedure starts with a model of the Allan variance for a particular oscillator and develops the statistics of range and integrated Doppler measurements. A series of five first order Markov processes is used to approximate the power spectral density obtained from the Allan variance.
Assessing Multivariate Constraints to Evolution across Ten Long-Term Avian Studies
Teplitsky, Celine; Tarka, Maja; Møller, Anders P.; Nakagawa, Shinichi; Balbontín, Javier; Burke, Terry A.; Doutrelant, Claire; Gregoire, Arnaud; Hansson, Bengt; Hasselquist, Dennis; Gustafsson, Lars; de Lope, Florentino; Marzal, Alfonso; Mills, James A.; Wheelwright, Nathaniel T.; Yarrall, John W.; Charmantier, Anne
2014-01-01
Background In a rapidly changing world, it is of fundamental importance to understand processes constraining or facilitating adaptation through microevolution. As different traits of an organism covary, genetic correlations are expected to affect evolutionary trajectories. However, only limited empirical data are available. Methodology/Principal Findings We investigate the extent to which multivariate constraints affect the rate of adaptation, focusing on four morphological traits often shown to harbour large amounts of genetic variance and considered to be subject to limited evolutionary constraints. Our data set includes unique long-term data for seven bird species and a total of 10 populations. We estimate population-specific matrices of genetic correlations and multivariate selection coefficients to predict evolutionary responses to selection. Using Bayesian methods that facilitate the propagation of errors in estimates, we compare (1) the rate of adaptation based on predicted response to selection when including genetic correlations with predictions from models where these genetic correlations were set to zero and (2) the multivariate evolvability in the direction of current selection to the average evolvability in random directions of the phenotypic space. We show that genetic correlations on average decrease the predicted rate of adaptation by 28%. Multivariate evolvability in the direction of current selection was systematically lower than average evolvability in random directions of space. These significant reductions in the rate of adaptation and reduced evolvability were due to a general nonalignment of selection and genetic variance, notably orthogonality of directional selection with the size axis along which most (60%) of the genetic variance is found. Conclusions These results suggest that genetic correlations can impose significant constraints on the evolution of avian morphology in wild populations. This could have important impacts on evolutionary dynamics and hence population persistence in the face of rapid environmental change. PMID:24608111
Fearon, Elizabeth; Chabata, Sungai T; Thompson, Jennifer A; Cowan, Frances M; Hargreaves, James R
2017-09-14
While guidance exists for obtaining population size estimates using multiplier methods with respondent-driven sampling surveys, we lack specific guidance for making sample size decisions. To guide the design of multiplier method population size estimation studies using respondent-driven sampling surveys to reduce the random error around the estimate obtained. The population size estimate is obtained by dividing the number of individuals receiving a service or the number of unique objects distributed (M) by the proportion of individuals in a representative survey who report receipt of the service or object (P). We have developed an approach to sample size calculation, interpreting methods to estimate the variance around estimates obtained using multiplier methods in conjunction with research into design effects and respondent-driven sampling. We describe an application to estimate the number of female sex workers in Harare, Zimbabwe. There is high variance in estimates. Random error around the size estimate reflects uncertainty from M and P, particularly when the estimate of P in the respondent-driven sampling survey is low. As expected, sample size requirements are higher when the design effect of the survey is assumed to be greater. We suggest a method for investigating the effects of sample size on the precision of a population size estimate obtained using multipler methods and respondent-driven sampling. Uncertainty in the size estimate is high, particularly when P is small, so balancing against other potential sources of bias, we advise researchers to consider longer service attendance reference periods and to distribute more unique objects, which is likely to result in a higher estimate of P in the respondent-driven sampling survey. ©Elizabeth Fearon, Sungai T Chabata, Jennifer A Thompson, Frances M Cowan, James R Hargreaves. Originally published in JMIR Public Health and Surveillance (http://publichealth.jmir.org), 14.09.2017.
NASA Astrophysics Data System (ADS)
Dohe, S.; Sherlock, V.; Hase, F.; Gisi, M.; Robinson, J.; Sepúlveda, E.; Schneider, M.; Blumenstock, T.
2013-08-01
The Total Carbon Column Observing Network (TCCON) has been established to provide ground-based remote sensing measurements of the column-averaged dry air mole fractions (DMF) of key greenhouse gases. To ensure network-wide consistency, biases between Fourier transform spectrometers at different sites have to be well controlled. Errors in interferogram sampling can introduce significant biases in retrievals. In this study we investigate a two-step scheme to correct these errors. In the first step the laser sampling error (LSE) is estimated by determining the sampling shift which minimises the magnitude of the signal intensity in selected, fully absorbed regions of the solar spectrum. The LSE is estimated for every day with measurements which meet certain selection criteria to derive the site-specific time series of the LSEs. In the second step, this sequence of LSEs is used to resample all the interferograms acquired at the site, and hence correct the sampling errors. Measurements acquired at the Izaña and Lauder TCCON sites are used to demonstrate the method. At both sites the sampling error histories show changes in LSE due to instrument interventions (e.g. realignment). Estimated LSEs are in good agreement with sampling errors inferred from the ratio of primary and ghost spectral signatures in optically bandpass-limited tungsten lamp spectra acquired at Lauder. The original time series of Xair and XCO2 (XY: column-averaged DMF of the target gas Y) at both sites show discrepancies of 0.2-0.5% due to changes in the LSE associated with instrument interventions or changes in the measurement sample rate. After resampling, discrepancies are reduced to 0.1% or less at Lauder and 0.2% at Izaña. In the latter case, coincident changes in interferometer alignment may also have contributed to the residual difference. In the future the proposed method will be used to correct historical spectra at all TCCON sites.
Measuring skewness of red blood cell deformability distribution by laser ektacytometry
DOE Office of Scientific and Technical Information (OSTI.GOV)
Nikitin, S Yu; Priezzhev, A V; Lugovtsov, A E
An algorithm is proposed for measuring the parameters of red blood cell deformability distribution based on laser diffractometry of red blood cells in shear flow (ektacytometry). The algorithm is tested on specially prepared samples of rat blood. In these experiments we succeeded in measuring the mean deformability, deformability variance and skewness of red blood cell deformability distribution with errors of 10%, 15% and 35%, respectively. (laser biophotonics)
Qu, Long; Guennel, Tobias; Marshall, Scott L
2013-12-01
Following the rapid development of genome-scale genotyping technologies, genetic association mapping has become a popular tool to detect genomic regions responsible for certain (disease) phenotypes, especially in early-phase pharmacogenomic studies with limited sample size. In response to such applications, a good association test needs to be (1) applicable to a wide range of possible genetic models, including, but not limited to, the presence of gene-by-environment or gene-by-gene interactions and non-linearity of a group of marker effects, (2) accurate in small samples, fast to compute on the genomic scale, and amenable to large scale multiple testing corrections, and (3) reasonably powerful to locate causal genomic regions. The kernel machine method represented in linear mixed models provides a viable solution by transforming the problem into testing the nullity of variance components. In this study, we consider score-based tests by choosing a statistic linear in the score function. When the model under the null hypothesis has only one error variance parameter, our test is exact in finite samples. When the null model has more than one variance parameter, we develop a new moment-based approximation that performs well in simulations. Through simulations and analysis of real data, we demonstrate that the new test possesses most of the aforementioned characteristics, especially when compared to existing quadratic score tests or restricted likelihood ratio tests. © 2013, The International Biometric Society.
Response Monitoring and Adjustment: Differential Relations with Psychopathic Traits
Bresin, Konrad; Finy, M. Sima; Sprague, Jenessa; Verona, Edelyn
2014-01-01
Studies on the relation between psychopathy and cognitive functioning often show mixed results, partially because different factors of psychopathy have not been considered fully. Based on previous research, we predicted divergent results based on a two-factor model of psychopathy (interpersonal-affective traits and impulsive-antisocial traits). Specifically, we predicted that the unique variance of interpersonal-affective traits would be related to increased monitoring (i.e., error-related negativity) and adjusting to errors (i.e., post-error slowing), whereas impulsive-antisocial traits would be related to reductions in these processes. Three studies using a diverse selection of assessment tools, samples, and methods are presented to identify response monitoring correlates of the two main factors of psychopathy. In Studies 1 (undergraduates), 2 (adolescents), and 3 (offenders), interpersonal-affective traits were related to increased adjustment following errors and, in Study 3, to enhanced monitoring of errors. Impulsive-antisocial traits were not consistently related to error adjustment across the studies, although these traits were related to a deficient monitoring of errors in Study 3. The results may help explain previous mixed findings and advance implications for etiological models of psychopathy. PMID:24933282
NASA Technical Reports Server (NTRS)
Fisher, Brad; Wolff, David B.
2010-01-01
Passive and active microwave rain sensors onboard earth-orbiting satellites estimate monthly rainfall from the instantaneous rain statistics collected during satellite overpasses. It is well known that climate-scale rain estimates from meteorological satellites incur sampling errors resulting from the process of discrete temporal sampling and statistical averaging. Sampling and retrieval errors ultimately become entangled in the estimation of the mean monthly rain rate. The sampling component of the error budget effectively introduces statistical noise into climate-scale rain estimates that obscure the error component associated with the instantaneous rain retrieval. Estimating the accuracy of the retrievals on monthly scales therefore necessitates a decomposition of the total error budget into sampling and retrieval error quantities. This paper presents results from a statistical evaluation of the sampling and retrieval errors for five different space-borne rain sensors on board nine orbiting satellites. Using an error decomposition methodology developed by one of the authors, sampling and retrieval errors were estimated at 0.25 resolution within 150 km of ground-based weather radars located at Kwajalein, Marshall Islands and Melbourne, Florida. Error and bias statistics were calculated according to the land, ocean and coast classifications of the surface terrain mask developed for the Goddard Profiling (GPROF) rain algorithm. Variations in the comparative error statistics are attributed to various factors related to differences in the swath geometry of each rain sensor, the orbital and instrument characteristics of the satellite and the regional climatology. The most significant result from this study found that each of the satellites incurred negative longterm oceanic retrieval biases of 10 to 30%.
Olivares, M; Larrañaga, A; Irazola, M; Sarmiento, A; Murelaga, X; Etxebarria, N
2012-08-30
The determination of crystal size of chert samples can provide suitable information about the raw material used for the manufacture of archeological items. X-ray diffraction (XRD) has been widely used for this purpose in several scientific areas. However, the historical value of archeological pieces makes this procedure sometimes unfeasible and thus, non-invasive new analytical approaches are required. In this sense, a new method was developed relating the crystal size obtained by means of XRD and infrared spectroscopy (IR) using partial least squares regression. The IR spectra collected from a large amount of different geological chert samples of archeological use were pre-processed following different treatments (i.e., derivatization or sample-wise normalization) to obtain the best regression model. The full cross-validation was satisfactorily validated using real samples and the experimental root mean standard error of precision value was 165 Å whereas the average precision of the estimated size value was 3%. The features of infrared bands were also evaluated in order to know the background of the prediction ability. In the studied case, the variance in the model was associated to the differences in the characteristic stretching and bending infrared bands of SiO(2). Based on this fact, it would be feasible to estimate the crystal size if it is built beforehand a chemometric model relating the size measured by standard methods and the IR spectra. Copyright © 2012 Elsevier B.V. All rights reserved.
Adaptive Resampling Particle Filters for GPS Carrier-Phase Navigation and Collision Avoidance System
NASA Astrophysics Data System (ADS)
Hwang, Soon Sik
This dissertation addresses three problems: 1) adaptive resampling technique (ART) for Particle Filters, 2) precise relative positioning using Global Positioning System (GPS) Carrier-Phase (CP) measurements applied to nonlinear integer resolution problem for GPS CP navigation using Particle Filters, and 3) collision detection system based on GPS CP broadcasts. First, Monte Carlo filters, called Particle Filters (PF), are widely used where the system is non-linear and non-Gaussian. In real-time applications, their estimation accuracies and efficiencies are significantly affected by the number of particles and the scheduling of relocating weights and samples, the so-called resampling step. In this dissertation, the appropriate number of particles is estimated adaptively such that the error of the sample mean and variance stay in bounds. These bounds are given by the confidence interval of a normal probability distribution for a multi-variate state. Two required number of samples maintaining the mean and variance error within the bounds are derived. The time of resampling is determined when the required sample number for the variance error crosses the required sample number for the mean error. Second, the PF using GPS CP measurements with adaptive resampling is applied to precise relative navigation between two GPS antennas. In order to make use of CP measurements for navigation, the unknown number of cycles between GPS antennas, the so called integer ambiguity, should be resolved. The PF is applied to this integer ambiguity resolution problem where the relative navigation states estimation involves nonlinear observations and nonlinear dynamics equation. Using the PF, the probability density function of the states is estimated by sampling from the position and velocity space and the integer ambiguities are resolved without using the usual hypothesis tests to search for the integer ambiguity. The ART manages the number of position samples and the frequency of the resampling step for real-time kinematics GPS navigation. The experimental results demonstrate the performance of the ART and the insensitivity of the proposed approach to GPS CP cycle-slips. Third, the GPS has great potential for the development of new collision avoidance systems and is being considered for the next generation Traffic alert and Collision Avoidance System (TCAS). The current TCAS equipment, is capable of broadcasting GPS code information to nearby airplanes, and also, the collision avoidance system using the navigation information based on GPS code has been studied by researchers. In this dissertation, the aircraft collision detection system using GPS CP information is addressed. The PF with position samples is employed for the CP based relative position estimation problem and the same algorithm can be used to determine the vehicle attitude if multiple GPS antennas are used. For a reliable and enhanced collision avoidance system, three dimensional trajectories are projected using the estimates of the relative position, velocity, and the attitude. It is shown that the performance of GPS CP based collision detecting algorithm meets the accuracy requirements for a precise approach of flight for auto landing with significantly less unnecessary collision false alarms and no miss alarms.
Numerical prediction of a draft tube flow taking into account uncertain inlet conditions
NASA Astrophysics Data System (ADS)
Brugiere, O.; Balarac, G.; Corre, C.; Metais, O.; Flores, E.; Pleroy
2012-11-01
The swirling turbulent flow in a hydroturbine draft tube is computed with a non-intrusive uncertainty quantification (UQ) method coupled to Reynolds-Averaged Navier-Stokes (RANS) modelling in order to take into account in the numerical prediction the physical uncertainties existing on the inlet flow conditions. The proposed approach yields not only mean velocity fields to be compared with measured profiles, as is customary in Computational Fluid Dynamics (CFD) practice, but also variance of these quantities from which error bars can be deduced on the computed profiles, thus making more significant the comparison between experiment and computation.
Estimation After a Group Sequential Trial.
Milanzi, Elasma; Molenberghs, Geert; Alonso, Ariel; Kenward, Michael G; Tsiatis, Anastasios A; Davidian, Marie; Verbeke, Geert
2015-10-01
Group sequential trials are one important instance of studies for which the sample size is not fixed a priori but rather takes one of a finite set of pre-specified values, dependent on the observed data. Much work has been devoted to the inferential consequences of this design feature. Molenberghs et al (2012) and Milanzi et al (2012) reviewed and extended the existing literature, focusing on a collection of seemingly disparate, but related, settings, namely completely random sample sizes, group sequential studies with deterministic and random stopping rules, incomplete data, and random cluster sizes. They showed that the ordinary sample average is a viable option for estimation following a group sequential trial, for a wide class of stopping rules and for random outcomes with a distribution in the exponential family. Their results are somewhat surprising in the sense that the sample average is not optimal, and further, there does not exist an optimal, or even, unbiased linear estimator. However, the sample average is asymptotically unbiased, both conditionally upon the observed sample size as well as marginalized over it. By exploiting ignorability they showed that the sample average is the conventional maximum likelihood estimator. They also showed that a conditional maximum likelihood estimator is finite sample unbiased, but is less efficient than the sample average and has the larger mean squared error. Asymptotically, the sample average and the conditional maximum likelihood estimator are equivalent. This previous work is restricted, however, to the situation in which the the random sample size can take only two values, N = n or N = 2 n . In this paper, we consider the more practically useful setting of sample sizes in a the finite set { n 1 , n 2 , …, n L }. It is shown that the sample average is then a justifiable estimator , in the sense that it follows from joint likelihood estimation, and it is consistent and asymptotically unbiased. We also show why simulations can give the false impression of bias in the sample average when considered conditional upon the sample size. The consequence is that no corrections need to be made to estimators following sequential trials. When small-sample bias is of concern, the conditional likelihood estimator provides a relatively straightforward modification to the sample average. Finally, it is shown that classical likelihood-based standard errors and confidence intervals can be applied, obviating the need for technical corrections.
NASA Astrophysics Data System (ADS)
Lee, Y.; Keehm, Y.
2011-12-01
Estimating the degree of weathering in stone cultural heritage, such as pagodas and statues is very important to plan conservation and restoration. The ultrasonic measurement is one of commonly-used techniques to evaluate weathering index of stone cultual properties, since it is easy to use and non-destructive. Typically we use a portable ultrasonic device, PUNDIT with exponential sensors. However, there are many factors to cause errors in measurements such as operators, sensor layouts or measurement directions. In this study, we carried out variety of measurements with different operators (male and female), different sensor layouts (direct and indirect), and sensor directions (anisotropy). For operators bias, we found that there were not significant differences by the operator's sex, while the pressure an operator exerts can create larger error in measurements. Calibrating with a standard sample for each operator is very essential in this case. For the sensor layout, we found that the indirect measurement (commonly used for cultural properties, since the direct measurement is difficult in most cases) gives lower velocity than the real one. We found that the correction coefficient is slightly different for different types of rocks: 1.50 for granite and sandstone and 1.46 for marble. From the sensor directions, we found that many rocks have slight anisotropy in their ultrasonic velocity measurement, though they are considered isotropic in macroscopic scale. Thus averaging four different directional measurement (0°, 45°, 90°, 135°) gives much less errors in measurements (the variance is 2-3 times smaller). In conclusion, we reported the error in ultrasonic meaurement of stone cultural properties by various sources quantitatively and suggested the amount of correction and procedures to calibrate the measurements. Acknowledgement: This study, which forms a part of the project, has been achieved with the support of national R&D project, which has been hosted by National Research Institute of Cultural Heritage of Cultural Heritage Administration(No. NRICH-1107-B01F).
A note on variance estimation in random effects meta-regression.
Sidik, Kurex; Jonkman, Jeffrey N
2005-01-01
For random effects meta-regression inference, variance estimation for the parameter estimates is discussed. Because estimated weights are used for meta-regression analysis in practice, the assumed or estimated covariance matrix used in meta-regression is not strictly correct, due to possible errors in estimating the weights. Therefore, this note investigates the use of a robust variance estimation approach for obtaining variances of the parameter estimates in random effects meta-regression inference. This method treats the assumed covariance matrix of the effect measure variables as a working covariance matrix. Using an example of meta-analysis data from clinical trials of a vaccine, the robust variance estimation approach is illustrated in comparison with two other methods of variance estimation. A simulation study is presented, comparing the three methods of variance estimation in terms of bias and coverage probability. We find that, despite the seeming suitability of the robust estimator for random effects meta-regression, the improved variance estimator of Knapp and Hartung (2003) yields the best performance among the three estimators, and thus may provide the best protection against errors in the estimated weights.
Clark, S; Rose, D J
2001-04-01
To establish reliability estimates of the 75% Limits of Stability Test (75% LOS test) when administered to community-dwelling older adults with a history of falls. Generalizability theory was used to estimate both the relative contribution of identified error sources to the total measurement error and generalizability coefficients. A random effects repeated-measures analysis of variance (ANOVA) was used to assess consistency of LOS test movement variables across both days and targets. A motor control research laboratory in a university setting. Fifty community-dwelling older adults with 2 or more falls in the previous year. Spatial and temporal measures of dynamic balance derived from the 75% LOS test included average movement velocity, maximum center of gravity (COG) excursion, end-point COG excursion, and directional control. Estimated generalizability coefficients for 2 testing days ranged from.58 to.87. Total variance in LOS test measures attributable to inconsistencies in day-to-day test performance (Day and Subject x Day facets) ranged from 2.5% to 8.4%. The ANOVA results indicated that no significant differences were observed in the LOS test variables across the 2 testing days. The 75% LOS test administered to older adult fallers on 2 consecutive days provides consistent and reliable measures of dynamic balance.
Jahng, Seungmin; Wood, Phillip K.
2017-01-01
Intensive longitudinal studies, such as ecological momentary assessment studies using electronic diaries, are gaining popularity across many areas of psychology. Multilevel models (MLMs) are most widely used analytical tools for intensive longitudinal data (ILD). Although ILD often have individually distinct patterns of serial correlation of measures over time, inferences of the fixed effects, and random components in MLMs are made under the assumption that all variance and autocovariance components are homogenous across individuals. In the present study, we introduced a multilevel model with Cholesky transformation to model ILD with individually heterogeneous covariance structure. In addition, the performance of the transformation method and the effects of misspecification of heterogeneous covariance structure were investigated through a Monte Carlo simulation. We found that, if individually heterogeneous covariances are incorrectly assumed as homogenous independent or homogenous autoregressive, MLMs produce highly biased estimates of the variance of random intercepts and the standard errors of the fixed intercept and the fixed effect of a level 2 covariate when the average autocorrelation is high. For intensive longitudinal data with individual specific residual covariance, the suggested transformation method showed lower bias in those estimates than the misspecified models when the number of repeated observations within individuals is 50 or more. PMID:28286490
Mortazavi, Forough; Mortazavi, Saideh S.; Khosrorad, Razieh
2015-01-01
Background: Procrastination is a common behavior which affects different aspects of life. The procrastination assessment scale-student (PASS) evaluates academic procrastination apropos its frequency and reasons. Objectives: The aims of the present study were to translate, culturally adapt, and validate the Farsi version of the PASS in a sample of Iranian medical students. Patients and Methods: In this cross-sectional study, the PASS was translated into Farsi through the forward-backward method, and its content validity was thereafter assessed by a panel of 10 experts. The Farsi version of the PASS was subsequently distributed among 423 medical students. The internal reliability of the PASS was assessed using Cronbach’s alpha. An exploratory factor analysis (EFA) was conducted on 18 items and then 28 items of the scale to find new models. The construct validity of the scale was assessed using both EFA and confirmatory factor analysis. The predictive validity of the scale was evaluated by calculating the correlation between the academic procrastination scores and the students’ average scores in the previous semester. Results: The corresponding reliability of the first and second parts of the scale was 0.781 and 0.861. An EFA on 18 items of the scale found 4 factors which jointly explained 53.2% of variances: The model was marginally acceptable (root mean square error of approximation [RMSEA] =0.098, standardized root mean square residual [SRMR] =0.076, χ2 /df =4.8, comparative fit index [CFI] =0.83). An EFA on 28 items of the scale found 4 factors which altogether explained 42.62% of variances: The model was acceptable (RMSEA =0.07, SRMR =0.07, χ2/df =2.8, incremental fit index =0.90, CFI =0.90). There was a negative correlation between the procrastination scores and the students’ average scores (r = -0.131, P =0.02). Conclusions: The Farsi version of the PASS is a valid and reliable tool to measure academic procrastination in Iranian undergraduate medical students. PMID:26473078
NASA Astrophysics Data System (ADS)
Suparman, Yusep; Folmer, Henk; Oud, Johan H. L.
2014-01-01
Omitted variables and measurement errors in explanatory variables frequently occur in hedonic price models. Ignoring these problems leads to biased estimators. In this paper, we develop a constrained autoregression-structural equation model (ASEM) to handle both types of problems. Standard panel data models to handle omitted variables bias are based on the assumption that the omitted variables are time-invariant. ASEM allows handling of both time-varying and time-invariant omitted variables by constrained autoregression. In the case of measurement error, standard approaches require additional external information which is usually difficult to obtain. ASEM exploits the fact that panel data are repeatedly measured which allows decomposing the variance of a variable into the true variance and the variance due to measurement error. We apply ASEM to estimate a hedonic housing model for urban Indonesia. To get insight into the consequences of measurement error and omitted variables, we compare the ASEM estimates with the outcomes of (1) a standard SEM, which does not account for omitted variables, (2) a constrained autoregression model, which does not account for measurement error, and (3) a fixed effects hedonic model, which ignores measurement error and time-varying omitted variables. The differences between the ASEM estimates and the outcomes of the three alternative approaches are substantial.
NASA Technical Reports Server (NTRS)
Alston, D. W.
1981-01-01
The considered research had the objective to design a statistical model that could perform an error analysis of curve fits of wind tunnel test data using analysis of variance and regression analysis techniques. Four related subproblems were defined, and by solving each of these a solution to the general research problem was obtained. The capabilities of the evolved true statistical model are considered. The least squares fit is used to determine the nature of the force, moment, and pressure data. The order of the curve fit is increased in order to delete the quadratic effect in the residuals. The analysis of variance is used to determine the magnitude and effect of the error factor associated with the experimental data.
The Impact of Truth Surrogate Variance on Quality Assessment/Assurance in Wind Tunnel Testing
NASA Technical Reports Server (NTRS)
DeLoach, Richard
2016-01-01
Minimum data volume requirements for wind tunnel testing are reviewed and shown to depend on error tolerance, response model complexity, random error variance in the measurement environment, and maximum acceptable levels of inference error risk. Distinctions are made between such related concepts as quality assurance and quality assessment in response surface modeling, as well as between precision and accuracy. Earlier research on the scaling of wind tunnel tests is extended to account for variance in the truth surrogates used at confirmation sites in the design space to validate proposed response models. A model adequacy metric is presented that represents the fraction of the design space within which model predictions can be expected to satisfy prescribed quality specifications. The impact of inference error on the assessment of response model residuals is reviewed. The number of sites where reasonably well-fitted response models actually predict inadequately is shown to be considerably less than the number of sites where residuals are out of tolerance. The significance of such inference error effects on common response model assessment strategies is examined.
NASA Technical Reports Server (NTRS)
Sidik, S. M.
1972-01-01
The error variance of the process prior multivariate normal distributions of the parameters of the models are assumed to be specified, prior probabilities of the models being correct. A rule for termination of sampling is proposed. Upon termination, the model with the largest posterior probability is chosen as correct. If sampling is not terminated, posterior probabilities of the models and posterior distributions of the parameters are computed. An experiment was chosen to maximize the expected Kullback-Leibler information function. Monte Carlo simulation experiments were performed to investigate large and small sample behavior of the sequential adaptive procedure.
Enhancement of MS2D Bartington point measurement of soil magnetic susceptibility
NASA Astrophysics Data System (ADS)
Fabijańczyk, Piotr; Zawadzki, Jarosław
2015-04-01
Field magnetometry is fast method used to assess the potential soil pollution. The most popular device used to measure the soil magnetic susceptibility on the soil surface is a MS2D Bartington. Single reading using MS2D device of soil magnetic susceptibility is low time-consuming but often characterized by considerable errors related to the instrument or environmental and lithogenic factors. Typically, in order to calculate the reliable average value of soil magnetic susceptibility, a series of MS2D readings is performed in the sample point. As it was analyzed previously, such methodology makes it possible to significantly reduce the nugget effect of the variograms of soil magnetic susceptibility that is related to the micro-scale variance and measurement errors. The goal of this study was to optimize the process of taking a series of MS2D readings, whose average value constitutes a single measurement, in order to take into account micro-scale variations of soil magnetic susceptibility in proper determination of this parameter. This was done using statistical and geostatistical analyses. The analyses were performed using field MS2D measurements that were carried out in the study area located in the direct vicinity of the Katowice agglomeration. At 150 sample points 10 MS2D readings of soil magnetic susceptibility were taken. Using this data set, series of experimental variograms were calculated and modeled. Firstly, using single random MS2D reading for each sample point, and next using the data set increased by adding one more MS2D reading, until their number reached 10. The parameters of variogram: nugget effect, sill and range of correlation were used to determine the most suitable number of MS2D readings at sample point. The distributions of soil magnetic susceptibility at sample point were also analyzed in order to determine adequate number of readings enabling to calculate reliable average soil magnetic susceptibility. The research leading to these results has received funding from the Polish-Norwegian Research Programme operated by the National Centre for Research and Development under the Norwegian Financial Mechanism 2009-2014 in the frame of Project IMPACT - Contract No Pol-Nor/199338/45/2013. References: Zawadzki J., Magiera T., Fabijańczyk P., 2007. The influence of forest stand and organic horizon development on soil surface measurement of magnetic susceptibility. Polish Journal of Soil Science, XL(2), 113-124 Zawadzki J., Fabijańczyk P., Magiera T., Strzyszcz Z., 2010. Study of litter influence on magnetic susceptibility measurements of urban forest topsoils using the MS2D sensor. Environmental Earth Sciences, 61(2), 223-230.
A method for determining the weak statistical stationarity of a random process
NASA Technical Reports Server (NTRS)
Sadeh, W. Z.; Koper, C. A., Jr.
1978-01-01
A method for determining the weak statistical stationarity of a random process is presented. The core of this testing procedure consists of generating an equivalent ensemble which approximates a true ensemble. Formation of an equivalent ensemble is accomplished through segmenting a sufficiently long time history of a random process into equal, finite, and statistically independent sample records. The weak statistical stationarity is ascertained based on the time invariance of the equivalent-ensemble averages. Comparison of these averages with their corresponding time averages over a single sample record leads to a heuristic estimate of the ergodicity of a random process. Specific variance tests are introduced for evaluating the statistical independence of the sample records, the time invariance of the equivalent-ensemble autocorrelations, and the ergodicity. Examination and substantiation of these procedures were conducted utilizing turbulent velocity signals.
1951-05-01
prccedur&:s to be of hipn accuracy. Ambij;uity of subject responizes due to overlap of entries on tU,, record sheets vas negligible. Handwriting ...experimental variables on reading errors us carried out by analysis of variance methods. For this purpose it was convenient to consider different classes...on any scale - an error ofY one numbered division. For this reason, the result. of the analysis of variance of the /10’s errors by dial types may
NASA Technical Reports Server (NTRS)
Perigaud, Claire; Delecluse, Pascale
1993-01-01
Sea level variations of the Indian Ocean north of 20 deg S are analyzed from Geosat satellite altimeter data over April 1985-September 1989. These variations are compared and interpreted with numerical simulations derived from a reduced gravity model forced by FSU observed winds over the same period. After decomposition into complex empirical orthogonal functions, the low-frequency anomalies are described by the first two modes for observations as well as for simulations. The sums of the two modes contain 34% and 40% of the observed and simulated variances, respectively. Averaged over the basin, the observed and simulated sea level changes are correlated by 0.92 over 1985-1988. The strongest change happens during the El Ninio 1986-1987: between winter 1986 and summer 1987 the basin-averaged sea level rises by approx. 1 cm. These low-frequency variations can partly be explained by changes in the Sverdrup circulation. The southern tropical Indian Ocean between 1O deg and 20 deg S is the domain where those changes are strongest: the averaged sea level rises by approx. 4.5 cm between winter 1986 and winter 1987. There, the signal propagates southwestward across the basin at a speed similar to free Rossby waves. Sensitivity of observed anomalies is examined over 1987-1988, with different orbit ephemeris, tropospheric corrections, and error reduction processes. The uncertainty of the basin-averaged sea level estimates is mostly due to the way the orbit error is reduced and reaches approx. 1 cm. Nonetheless, spatial correlation is good between the various observations and better than between observations and simulations. Sensitivity of simulated anomalies to the wind uncertainty, examined with Former Soviet Union (FSU) and European Center for Medium-Range Weather Forecasting (ECMWF) forcings over 1985-1988, shows that the variance of the simulations driven by ECMWF is 52% smaller, as FSU winds are stronger than ECMWF. Results show that the wind strength also affects the dynamic response of the ocean: anomalies propagate westward across the basin more than twice as fast with FSU than with ECMWF. It is found that the discrepancy is larger between ECMWF and FSU simulations than between observations and FSU simulations.
Yu, Shaohui; Xiao, Xue; Ding, Hong; Xu, Ge; Li, Haixia; Liu, Jing
2017-08-05
The quantitative analysis is very difficult for the emission-excitation fluorescence spectroscopy of multi-component mixtures whose fluorescence peaks are serious overlapping. As an effective method for the quantitative analysis, partial least squares can extract the latent variables from both the independent variables and the dependent variables, so it can model for multiple correlations between variables. However, there are some factors that usually affect the prediction results of partial least squares, such as the noise, the distribution and amount of the samples in calibration set etc. This work focuses on the problems in the calibration set that are mentioned above. Firstly, the outliers in the calibration set are removed by leave-one-out cross-validation. Then, according to two different prediction requirements, the EWPLS method and the VWPLS method are proposed. The independent variables and dependent variables are weighted in the EWPLS method by the maximum error of the recovery rate and weighted in the VWPLS method by the maximum variance of the recovery rate. Three organic matters with serious overlapping excitation-emission fluorescence spectroscopy are selected for the experiments. The step adjustment parameter, the iteration number and the sample amount in the calibration set are discussed. The results show the EWPLS method and the VWPLS method are superior to the PLS method especially for the case of small samples in the calibration set. Copyright © 2017 Elsevier B.V. All rights reserved.
Teo, Troy P; Ahmed, Syed Bilal; Kawalec, Philip; Alayoubi, Nadia; Bruce, Neil; Lyn, Ethan; Pistorius, Stephen
2018-02-01
The accurate prediction of intrafraction lung tumor motion is required to compensate for system latency in image-guided adaptive radiotherapy systems. The goal of this study was to identify an optimal prediction model that has a short learning period so that prediction and adaptation can commence soon after treatment begins, and requires minimal reoptimization for individual patients. Specifically, the feasibility of predicting tumor position using a combination of a generalized (i.e., averaged) neural network, optimized using historical patient data (i.e., tumor trajectories) obtained offline, coupled with the use of real-time online tumor positions (obtained during treatment delivery) was examined. A 3-layer perceptron neural network was implemented to predict tumor motion for a prediction horizon of 650 ms. A backpropagation algorithm and batch gradient descent approach were used to train the model. Twenty-seven 1-min lung tumor motion samples (selected from a CyberKnife patient dataset) were sampled at a rate of 7.5 Hz (0.133 s) to emulate the frame rate of an electronic portal imaging device (EPID). A sliding temporal window was used to sample the data for learning. The sliding window length was set to be equivalent to the first breathing cycle detected from each trajectory. Performing a parametric sweep, an averaged error surface of mean square errors (MSE) was obtained from the prediction responses of seven trajectories used for the training of the model (Group 1). An optimal input data size and number of hidden neurons were selected to represent the generalized model. To evaluate the prediction performance of the generalized model on unseen data, twenty tumor traces (Group 2) that were not involved in the training of the model were used for the leave-one-out cross-validation purposes. An input data size of 35 samples (4.6 s) and 20 hidden neurons were selected for the generalized neural network. An average sliding window length of 28 data samples was used. The average initial learning period prior to the availability of the first predicted tumor position was 8.53 ± 1.03 s. Average mean absolute error (MAE) of 0.59 ± 0.13 mm and 0.56 ± 0.18 mm were obtained from Groups 1 and 2, respectively, giving an overall MAE of 0.57 ± 0.17 mm. Average root-mean-square-error (RMSE) of 0.67 ± 0.36 for all the traces (0.76 ± 0.34 mm, Group 1 and 0.63 ± 0.36 mm, Group 2), is comparable to previously published results. Prediction errors are mainly due to the irregular periodicities between cycles. Since the errors from Groups 1 and 2 are within the same range, it demonstrates that this model can generalize and predict on unseen data. This is a first attempt to use an averaged MSE error surface (obtained from the prediction of different patients' tumor trajectories) to determine the parameters of a generalized neural network. This network could be deployed as a plug-and-play predictor for tumor trajectory during treatment delivery, eliminating the need for optimizing individual networks with pretreatment patient data. © 2017 American Association of Physicists in Medicine.
Gray, Brian R.; Gitzen, Robert A.; Millspaugh, Joshua J.; Cooper, Andrew B.; Licht, Daniel S.
2012-01-01
Variance components may play multiple roles (cf. Cox and Solomon 2003). First, magnitudes and relative magnitudes of the variances of random factors may have important scientific and management value in their own right. For example, variation in levels of invasive vegetation among and within lakes may suggest causal agents that operate at both spatial scales – a finding that may be important for scientific and management reasons. Second, variance components may also be of interest when they affect precision of means and covariate coefficients. For example, variation in the effect of water depth on the probability of aquatic plant presence in a study of multiple lakes may vary by lake. This variation will affect the precision of the average depth-presence association. Third, variance component estimates may be used when designing studies, including monitoring programs. For example, to estimate the numbers of years and of samples per year required to meet long-term monitoring goals, investigators need estimates of within and among-year variances. Other chapters in this volume (Chapters 7, 8, and 10) as well as extensive external literature outline a framework for applying estimates of variance components to the design of monitoring efforts. For example, a series of papers with an ecological monitoring theme examined the relative importance of multiple sources of variation, including variation in means among sites, years, and site-years, for the purposes of temporal trend detection and estimation (Larsen et al. 2004, and references therein).
Structural parameters of young star clusters: fractal analysis
NASA Astrophysics Data System (ADS)
Hetem, A.
2017-07-01
A unified view of star formation in the Universe demand detailed and in-depth studies of young star clusters. This work is related to our previous study of fractal statistics estimated for a sample of young stellar clusters (Gregorio-Hetem et al. 2015, MNRAS 448, 2504). The structural properties can lead to significant conclusions about the early stages of cluster formation: 1) virial conditions can be used to distinguish warm collapsed; 2) bound or unbound behaviour can lead to conclusions about expansion; and 3) fractal statistics are correlated to the dynamical evolution and age. The technique of error bars estimation most used in the literature is to adopt inferential methods (like bootstrap) to estimate deviation and variance, which are valid only for an artificially generated cluster. In this paper, we expanded the number of studied clusters, in order to enhance the investigation of the cluster properties and dynamic evolution. The structural parameters were compared with fractal statistics and reveal that the clusters radial density profile show a tendency of the mean separation of the stars increase with the average surface density. The sample can be divided into two groups showing different dynamic behaviour, but they have the same dynamic evolution, since the entire sample was revealed as being expanding objects, for which the substructures do not seem to have been completely erased. These results are in agreement with the simulations adopting low surface densities and supervirial conditions.
Development and Initial Validation of the Multicultural Personality Inventory (MPI).
Ponterotto, Joseph G; Fietzer, Alexander W; Fingerhut, Esther C; Woerner, Scott; Stack, Lauren; Magaldi-Dopman, Danielle; Rust, Jonathan; Nakao, Gen; Tsai, Yu-Ting; Black, Natasha; Alba, Renaldo; Desai, Miraj; Frazier, Chantel; LaRue, Alyse; Liao, Pei-Wen
2014-01-01
Two studies summarize the development and initial validation of the Multicultural Personality Inventory (MPI). In Study 1, the 115-item prototype MPI was administered to 415 university students where exploratory factor analysis resulted in a 70-item, 7-factor model. In Study 2, the 70-item MPI and theoretically related companion instruments were administered to a multisite sample of 576 university students. Confirmatory factory analysis found the 7-factor structure to be a relatively good fit to the data (Comparative Fit Index =.954; root mean square error of approximation =.057), and MPI factors predicted variance in criterion variables above and beyond the variance accounted for by broad personality traits (i.e., Big Five). Study limitations and directions for further validation research are specified.
Ben Natan, Merav; Sharon, Ira; Mahajna, Marlen; Mahajna, Sara
2017-11-01
Medication errors are common among nursing students. Nonetheless, these errors are often underreported. To examine factors related to nursing students' intention to report medication errors, using the Theory of Planned Behavior, and to examine whether the theory is useful in predicting students' intention to report errors. This study has a descriptive cross-sectional design. Study population was recruited in a university and a large nursing school in central and northern Israel. A convenience sample of 250 nursing students took part in the study. The students completed a self-report questionnaire, based on the Theory of Planned Behavior. The findings indicate that students' intention to report medication errors was high. The Theory of Planned Behavior constructs explained 38% of variance in students' intention to report medication errors. The constructs of behavioral beliefs, subjective norms, and perceived behavioral control were found as affecting this intention, while the most significant factor was behavioral beliefs. The findings also reveal that students' fear of the reaction to disclosure of the error from superiors and colleagues may impede them from reporting the error. Understanding factors related to reporting medication errors is crucial to designing interventions that foster error reporting. Copyright © 2017 Elsevier Ltd. All rights reserved.
ERIC Educational Resources Information Center
Brockmann, Frank
2011-01-01
State testing programs today are more extensive than ever, and their results are required to serve more purposes and high-stakes decisions than one might have imagined. Assessment results are used to hold schools, districts, and states accountable for student performance and to help guide a multitude of important decisions. This report describes…
NASA Astrophysics Data System (ADS)
Behnabian, Behzad; Mashhadi Hossainali, Masoud; Malekzadeh, Ahad
2018-02-01
The cross-validation technique is a popular method to assess and improve the quality of prediction by least squares collocation (LSC). We present a formula for direct estimation of the vector of cross-validation errors (CVEs) in LSC which is much faster than element-wise CVE computation. We show that a quadratic form of CVEs follows Chi-squared distribution. Furthermore, a posteriori noise variance factor is derived by the quadratic form of CVEs. In order to detect blunders in the observations, estimated standardized CVE is proposed as the test statistic which can be applied when noise variances are known or unknown. We use LSC together with the methods proposed in this research for interpolation of crustal subsidence in the northern coast of the Gulf of Mexico. The results show that after detection and removing outliers, the root mean square (RMS) of CVEs and estimated noise standard deviation are reduced about 51 and 59%, respectively. In addition, RMS of LSC prediction error at data points and RMS of estimated noise of observations are decreased by 39 and 67%, respectively. However, RMS of LSC prediction error on a regular grid of interpolation points covering the area is only reduced about 4% which is a consequence of sparse distribution of data points for this case study. The influence of gross errors on LSC prediction results is also investigated by lower cutoff CVEs. It is indicated that after elimination of outliers, RMS of this type of errors is also reduced by 19.5% for a 5 km radius of vicinity. We propose a method using standardized CVEs for classification of dataset into three groups with presumed different noise variances. The noise variance components for each of the groups are estimated using restricted maximum-likelihood method via Fisher scoring technique. Finally, LSC assessment measures were computed for the estimated heterogeneous noise variance model and compared with those of the homogeneous model. The advantage of the proposed method is the reduction in estimated noise levels for those groups with the fewer number of noisy data points.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Garcia, O. E., E-mail: odd.erik.garcia@uit.no; Kube, R.; Theodorsen, A.
A stochastic model is presented for intermittent fluctuations in the scrape-off layer of magnetically confined plasmas. The fluctuations in the plasma density are modeled by a super-position of uncorrelated pulses with fixed shape and duration, describing radial motion of blob-like structures. In the case of an exponential pulse shape and exponentially distributed pulse amplitudes, predictions are given for the lowest order moments, probability density function, auto-correlation function, level crossings, and average times for periods spent above and below a given threshold level. Also, the mean squared errors on estimators of sample mean and variance for realizations of the process bymore » finite time series are obtained. These results are discussed in the context of single-point measurements of fluctuations in the scrape-off layer, broad density profiles, and implications for plasma–wall interactions due to the transient transport events in fusion grade plasmas. The results may also have wide applications for modelling fluctuations in other magnetized plasmas such as basic laboratory experiments and ionospheric irregularities.« less
Podsakoff, P M; MacKenzie, S B; Bommer, W H
1996-08-01
A meta-analysis was conducted to estimate more accurately the bivariate relationships between leadership behaviors, substitutes for leadership, and subordinate attitudes, and role perceptions and performance, and to examine the relative strengths of the relationships between these variables. Estimates of 435 relationships were obtained from 22 studies containing 36 independent samples. The findings showed that the combination of the substitutes variables and leader behaviors account for a majority of the variance in employee attitudes and role perceptions and a substantial proportion of the variance in in-role and extra-role performance; on average, the substitutes for leadership uniquely accounted for more of the variance in the criterion variables than did leader behaviors.
Impact of source collinearity in simulated PM 2.5 data on the PMF receptor model solution
NASA Astrophysics Data System (ADS)
Habre, Rima; Coull, Brent; Koutrakis, Petros
2011-12-01
Positive Matrix Factorization (PMF) is a factor analytic model used to identify particle sources and to estimate their contributions to PM 2.5 concentrations observed at receptor sites. Collinearity in source contributions due to meteorological conditions introduces uncertainty in the PMF solution. We simulated datasets of speciated PM 2.5 concentrations associated with three ambient particle sources: "Motor Vehicle" (MV), "Sodium Chloride" (NaCl), and "Sulfur" (S), and we varied the correlation structure between their mass contributions to simulate collinearity. We analyzed the datasets in PMF using the ME-2 multilinear engine. The Pearson correlation coefficients between the simulated and PMF-predicted source contributions and profiles are denoted by " G correlation" and " F correlation", respectively. In sensitivity analyses, we examined how the means or variances of the source contributions affected the stability of the PMF solution with collinearity. The % errors in predicting the average source contributions were 23, 80 and 23% for MV, NaCl, and S, respectively. On average, the NaCl contribution was overestimated, while MV and S contributions were underestimated. The ability of PMF to predict the contributions and profiles of the three sources deteriorated significantly as collinearity in their contributions increased. When the mean of NaCl or variance of NaCl and MV source contributions was increased, the deterioration in G correlation with increasing collinearity became less significant, and the ability of PMF to predict the NaCl and MV loading profiles improved. When the three factor profiles were simulated to share more elements, the decrease in G and F correlations became non-significant. Our findings agree with previous simulation studies reporting that correlated sources are predicted with higher error and bias. Consequently, the power to detect significant concentration-response estimates in health effect analyses weakens.
Ristić-Djurović, Jasna L; Ćirković, Saša; Mladenović, Pavle; Romčević, Nebojša; Trbovich, Alexander M
2018-04-01
A rough estimate indicated that use of samples of size not larger than ten is not uncommon in biomedical research and that many of such studies are limited to strong effects due to sample sizes smaller than six. For data collected from biomedical experiments it is also often unknown if mathematical requirements incorporated in the sample comparison methods are satisfied. Computer simulated experiments were used to examine performance of methods for qualitative sample comparison and its dependence on the effectiveness of exposure, effect intensity, distribution of studied parameter values in the population, and sample size. The Type I and Type II errors, their average, as well as the maximal errors were considered. The sample size 9 and the t-test method with p = 5% ensured error smaller than 5% even for weak effects. For sample sizes 6-8 the same method enabled detection of weak effects with errors smaller than 20%. If the sample sizes were 3-5, weak effects could not be detected with an acceptable error; however, the smallest maximal error in the most general case that includes weak effects is granted by the standard error of the mean method. The increase of sample size from 5 to 9 led to seven times more accurate detection of weak effects. Strong effects were detected regardless of the sample size and method used. The minimal recommended sample size for biomedical experiments is 9. Use of smaller sizes and the method of their comparison should be justified by the objective of the experiment. Copyright © 2018 Elsevier B.V. All rights reserved.
Characterizing error distributions for MISR and MODIS optical depth data
NASA Astrophysics Data System (ADS)
Paradise, S.; Braverman, A.; Kahn, R.; Wilson, B.
2008-12-01
The Multi-angle Imaging SpectroRadiometer (MISR) and Moderate Resolution Imaging Spectroradiometer (MODIS) on NASA's EOS satellites collect massive, long term data records on aerosol amounts and particle properties. MISR and MODIS have different but complementary sampling characteristics. In order to realize maximum scientific benefit from these data, the nature of their error distributions must be quantified and understood so that discrepancies between them can be rectified and their information combined in the most beneficial way. By 'error' we mean all sources of discrepancies between the true value of the quantity of interest and the measured value, including instrument measurement errors, artifacts of retrieval algorithms, and differential spatial and temporal sampling characteristics. Previously in [Paradise et al., Fall AGU 2007: A12A-05] we presented a unified, global analysis and comparison of MISR and MODIS measurement biases and variances over lives of the missions. We used AErosol RObotic NETwork (AERONET) data as ground truth and evaluated MISR and MODIS optical depth distributions relative to AERONET using simple linear regression. However, AERONET data are themselves instrumental measurements subject to sources of uncertainty. In this talk, we discuss results from an improved analysis of MISR and MODIS error distributions that uses errors-in-variables regression, accounting for uncertainties in both the dependent and independent variables. We demonstrate on optical depth data, but the method is generally applicable to other aerosol properties as well.
Interlaboratory comparison of red-cell ATP, 2,3-diphosphoglycerate and haemolysis measurements.
Hess, J R; Kagen, L R; van der Meer, P F; Simon, T; Cardigan, R; Greenwalt, T J; AuBuchon, J P; Brand, A; Lockwood, W; Zanella, A; Adamson, J; Snyder, E; Taylor, H L; Moroff, G; Hogman, C
2005-07-01
Red blood cell (RBC) storage systems are licensed based on their ability to prevent haemolysis and maintain RBC 24-h in vivo recovery. Preclinical testing includes measurement of RBC ATP as a surrogate for recovery, 2,3-diphosphoglycerate (DPG) as a surrogate for oxygen affinity, and free haemoglobin, which is indicative of red cell lysis. The reproducibility of RBC ATP, DPG and haemolysis measurements between centres was investigated. Five, 4-day-old leucoreduced AS-1 RBC units were pooled, aliquotted and shipped on ice to 14 laboratories in the USA and European Union (EU). Each laboratory was to sample the bag twice on day 7 and measure RBC ATP, DPG, haemoglobin and haemolysis levels in triplicate on each sample. The variability of results was assessed by using coefficients of variation (CV) and analysis of variance. Measurements were highly reproducible at the individual sites. Between sites, the CV was 16% for ATP, 35% for DPG, 2% for total haemoglobin and 54% for haemolysis. For ATP and total haemoglobin, 94 and 80% of the variance in measurements was contributed by differences between sites, and more than 80% of the variance for DPG and haemolysis measurements came from markedly discordant results from three sites and one site, respectively. In descending order, mathematical errors, unvalidated analytical methods, a lack of shared standards and fluid handling errors contributed to the variability in measurements from different sites. While the methods used by laboratories engaged in RBC storage system clinical trials demonstrated good precision, differences in results between laboratories may hinder comparative analysis. Efforts to improve performance should focus on developing robust methods, especially for measuring RBC ATP.
Assessment of averaging spatially correlated noise for 3-D radial imaging.
Stobbe, Robert W; Beaulieu, Christian
2011-07-01
Any measurement of signal intensity obtained from an image will be corrupted by noise. If the measurement is from one voxel, an error bound associated with noise can be assigned if the standard deviation of noise in the image is known. If voxels are averaged together within a region of interest (ROI) and the image noise is uncorrelated, the error bound associated with noise will be reduced in proportion to the square root of the number of voxels in the ROI. However, when 3-D-radial images are created the image noise will be spatially correlated. In this paper, an equation is derived and verified with simulated noise for the computation of noise averaging when image noise is correlated, facilitating the assessment of noise characteristics for different 3-D-radial imaging methodologies. It is already known that if the radial evolution of projections are altered such that constant sampling density is produced in k-space, the signal-to-noise ratio (SNR) inefficiency of standard radial imaging (SR) can effectively be eliminated (assuming a uniform transfer function is desired). However, it is shown in this paper that the low-frequency noise power reduction of SR will produce beneficial (anti-) correlation of noise and enhanced noise averaging characteristics. If an ROI contains only one voxel a radial evolution altered uniform k-space sampling technique such as twisted projection imaging (TPI) will produce an error bound ~35% less with respect to noise than SR, however, for an ROI containing 16 voxels the SR methodology will facilitate an error bound ~20% less than TPI. If a filtering transfer function is desired, it is shown that designing sampling density to create the filter shape has both SNR and noise correlation advantages over sampling k-space uniformly. In this context SR is also beneficial. Two sets of 48 images produced from a saline phantom with sodium MRI at 4.7T are used to experimentally measure noise averaging characteristics of radial imaging and good agreement with theory is obtained.
Mcalister, Courtney; Schmitter-Edgecombe, Maureen; Lamb, Richard
2016-01-01
The objective of this meta-analysis was to improve understanding of the heterogeneity in the relationship between cognition and functional status in individuals with mild cognitive impairment (MCI). Demographic, clinical, and methodological moderators were examined. Cognition explained an average of 23% of the variance in functional outcomes. Executive function measures explained the largest amount of variance (37%), whereas global cognitive status and processing speed measures explained the least (20%). Short- and long-delayed memory measures accounted for more variance (35% and 31%) than immediate memory measures (18%), and the relationship between cognition and functional outcomes was stronger when assessed with informant-report (28%) compared with self-report (21%). Demographics, sample characteristics, and type of everyday functioning measures (i.e., questionnaire, performance-based) explained relatively little variance compared with cognition. Executive functioning, particularly measured by Trails B, was a strong predictor of everyday functioning in individuals with MCI. A large proportion of variance remained unexplained by cognition. PMID:26743326
Statistically Self-Consistent and Accurate Errors for SuperDARN Data
NASA Astrophysics Data System (ADS)
Reimer, A. S.; Hussey, G. C.; McWilliams, K. A.
2018-01-01
The Super Dual Auroral Radar Network (SuperDARN)-fitted data products (e.g., spectral width and velocity) are produced using weighted least squares fitting. We present a new First-Principles Fitting Methodology (FPFM) that utilizes the first-principles approach of Reimer et al. (2016) to estimate the variance of the real and imaginary components of the mean autocorrelation functions (ACFs) lags. SuperDARN ACFs fitted by the FPFM do not use ad hoc or empirical criteria. Currently, the weighting used to fit the ACF lags is derived from ad hoc estimates of the ACF lag variance. Additionally, an overcautious lag filtering criterion is used that sometimes discards data that contains useful information. In low signal-to-noise (SNR) and/or low signal-to-clutter regimes the ad hoc variance and empirical criterion lead to underestimated errors for the fitted parameter because the relative contributions of signal, noise, and clutter to the ACF variance is not taken into consideration. The FPFM variance expressions include contributions of signal, noise, and clutter. The clutter is estimated using the maximal power-based self-clutter estimator derived by Reimer and Hussey (2015). The FPFM was successfully implemented and tested using synthetic ACFs generated with the radar data simulator of Ribeiro, Ponomarenko, et al. (2013). The fitted parameters and the fitted-parameter errors produced by the FPFM are compared with the current SuperDARN fitting software, FITACF. Using self-consistent statistical analysis, the FPFM produces reliable or trustworthy quantitative measures of the errors of the fitted parameters. For an SNR in excess of 3 dB and velocity error below 100 m/s, the FPFM produces 52% more data points than FITACF.
NASA Technical Reports Server (NTRS)
Bell, Thomas L.; Lau, William K. M. (Technical Monitor)
2002-01-01
The proposed Global Precipitation Mission (GPM) builds on the success of the Tropical Rainfall Measuring Mission (TRMM), offering a constellation of microwave-sensor-equipped smaller satellites in addition to a larger, multiply-instrumented "mother" satellite that will include an improved precipitation radar system to which the precipitation estimates of the smaller satellites can be tuned. Coverage by the satellites will be nearly global rather than being confined as TRMM was to lower latitudes. It is hoped that the satellite constellation can provide observations at most places on the earth at least once every three hours, though practical considerations may force some compromises. The GPM system offers the possibility of providing precipitation maps with much better time resolution than the monthly averages around which TRMM was planned, and therefore opens up new possibilities for hydrology and data assimilation into models. In this talk, methods that were developed for estimating sampling error in the rainfall averages that TRMM is providing will be used to estimate sampling error levels for GPM-era configurations. Possible impacts on GPM products of compromises in the sampling frequency will be discussed.
Accounting for dropout bias using mixed-effects models.
Mallinckrodt, C H; Clark, W S; David, S R
2001-01-01
Treatment effects are often evaluated by comparing change over time in outcome measures. However, valid analyses of longitudinal data can be problematic when subjects discontinue (dropout) prior to completing the study. This study assessed the merits of likelihood-based repeated measures analyses (MMRM) compared with fixed-effects analysis of variance where missing values were imputed using the last observation carried forward approach (LOCF) in accounting for dropout bias. Comparisons were made in simulated data and in data from a randomized clinical trial. Subject dropout was introduced in the simulated data to generate ignorable and nonignorable missingness. Estimates of treatment group differences in mean change from baseline to endpoint from MMRM were, on average, markedly closer to the true value than estimates from LOCF in every scenario simulated. Standard errors and confidence intervals from MMRM accurately reflected the uncertainty of the estimates, whereas standard errors and confidence intervals from LOCF underestimated uncertainty.
Guo, Jiin-Huarng; Luh, Wei-Ming
2009-05-01
When planning a study, sample size determination is one of the most important tasks facing the researcher. The size will depend on the purpose of the study, the cost limitations, and the nature of the data. By specifying the standard deviation ratio and/or the sample size ratio, the present study considers the problem of heterogeneous variances and non-normality for Yuen's two-group test and develops sample size formulas to minimize the total cost or maximize the power of the test. For a given power, the sample size allocation ratio can be manipulated so that the proposed formulas can minimize the total cost, the total sample size, or the sum of total sample size and total cost. On the other hand, for a given total cost, the optimum sample size allocation ratio can maximize the statistical power of the test. After the sample size is determined, the present simulation applies Yuen's test to the sample generated, and then the procedure is validated in terms of Type I errors and power. Simulation results show that the proposed formulas can control Type I errors and achieve the desired power under the various conditions specified. Finally, the implications for determining sample sizes in experimental studies and future research are discussed.
Hosseini, Marjan; Kerachian, Reza
2017-09-01
This paper presents a new methodology for analyzing the spatiotemporal variability of water table levels and redesigning a groundwater level monitoring network (GLMN) using the Bayesian Maximum Entropy (BME) technique and a multi-criteria decision-making approach based on ordered weighted averaging (OWA). The spatial sampling is determined using a hexagonal gridding pattern and a new method, which is proposed to assign a removal priority number to each pre-existing station. To design temporal sampling, a new approach is also applied to consider uncertainty caused by lack of information. In this approach, different time lag values are tested by regarding another source of information, which is simulation result of a numerical groundwater flow model. Furthermore, to incorporate the existing uncertainties in available monitoring data, the flexibility of the BME interpolation technique is taken into account in applying soft data and improving the accuracy of the calculations. To examine the methodology, it is applied to the Dehgolan plain in northwestern Iran. Based on the results, a configuration of 33 monitoring stations for a regular hexagonal grid of side length 3600 m is proposed, in which the time lag between samples is equal to 5 weeks. Since the variance estimation errors of the BME method are almost identical for redesigned and existing networks, the redesigned monitoring network is more cost-effective and efficient than the existing monitoring network with 52 stations and monthly sampling frequency.
Yi, Dong-Hoon; Lee, Tae-Jae; Cho, Dong-Il Dan
2015-05-13
This paper introduces a novel afocal optical flow sensor (OFS) system for odometry estimation in indoor robotic navigation. The OFS used in computer optical mouse has been adopted for mobile robots because it is not affected by wheel slippage. Vertical height variance is thought to be a dominant factor in systematic error when estimating moving distances in mobile robots driving on uneven surfaces. We propose an approach to mitigate this error by using an afocal (infinite effective focal length) system. We conducted experiments in a linear guide on carpet and three other materials with varying sensor heights from 30 to 50 mm and a moving distance of 80 cm. The same experiments were repeated 10 times. For the proposed afocal OFS module, a 1 mm change in sensor height induces a 0.1% systematic error; for comparison, the error for a conventional fixed-focal-length OFS module is 14.7%. Finally, the proposed afocal OFS module was installed on a mobile robot and tested 10 times on a carpet for distances of 1 m. The average distance estimation error and standard deviation are 0.02% and 17.6%, respectively, whereas those for a conventional OFS module are 4.09% and 25.7%, respectively.
Fixed precision sampling plans for white apple leafhopper (Homoptera: Cicadellidae) on apple.
Beers, Elizabeth H; Jones, Vincent P
2004-10-01
Constant precision sampling plans for the white apple leafhopper, Typhlocyba pomaria McAtee, were developed so that it could be used as an indicator species for system stability as new integrated pest management programs without broad-spectrum pesticides are developed. Taylor's power law was used to model the relationship between the mean and the variance, and Green's constant precision sequential sample equation was used to develop sampling plans. Bootstrap simulations of the sampling plans showed greater precision (D = 0.25) than the desired precision (Do = 0.3), particularly at low mean population densities. We found that by adjusting the Do value in Green's equation to 0.4, we were able to reduce the average sample number by 25% and provided an average D = 0.31. The sampling plan described allows T. pomaria to be used as reasonable indicator species of agroecosystem stability in Washington apple orchards.
FMRI group analysis combining effect estimates and their variances
Chen, Gang; Saad, Ziad S.; Nath, Audrey R.; Beauchamp, Michael S.; Cox, Robert W.
2012-01-01
Conventional functional magnetic resonance imaging (FMRI) group analysis makes two key assumptions that are not always justified. First, the data from each subject is condensed into a single number per voxel, under the assumption that within-subject variance for the effect of interest is the same across all subjects or is negligible relative to the cross-subject variance. Second, it is assumed that all data values are drawn from the same Gaussian distribution with no outliers. We propose an approach that does not make such strong assumptions, and present a computationally efficient frequentist approach to FMRI group analysis, which we term mixed-effects multilevel analysis (MEMA), that incorporates both the variability across subjects and the precision estimate of each effect of interest from individual subject analyses. On average, the more accurate tests result in higher statistical power, especially when conventional variance assumptions do not hold, or in the presence of outliers. In addition, various heterogeneity measures are available with MEMA that may assist the investigator in further improving the modeling. Our method allows group effect t-tests and comparisons among conditions and among groups. In addition, it has the capability to incorporate subject-specific covariates such as age, IQ, or behavioral data. Simulations were performed to illustrate power comparisons and the capability of controlling type I errors among various significance testing methods, and the results indicated that the testing statistic we adopted struck a good balance between power gain and type I error control. Our approach is instantiated in an open-source, freely distributed program that may be used on any dataset stored in the universal neuroimaging file transfer (NIfTI) format. To date, the main impediment for more accurate testing that incorporates both within- and cross-subject variability has been the high computational cost. Our efficient implementation makes this approach practical. We recommend its use in lieu of the less accurate approach in the conventional group analysis. PMID:22245637
Improved Event Location Uncertainty Estimates
2008-06-30
throughout this study . The data set consists of GT0-2 nuclear explosions from the SAIC Nuclear Explosion Database (www.rdss.info, Bahavar et al...errors: Bias and variance In this study SNR dependence of both delay and variance of reading errors of first arriving P waves are analyzed and...ground truth and range of event size. For other datasets we turn to estimates based on double- differences between arrival times of station pairs
Estimation and Simulation of Slow Crack Growth Parameters from Constant Stress Rate Data
NASA Technical Reports Server (NTRS)
Salem, Jonathan A.; Weaver, Aaron S.
2003-01-01
Closed form, approximate functions for estimating the variances and degrees-of-freedom associated with the slow crack growth parameters n, D, B, and A(sup *) as measured using constant stress rate ('dynamic fatigue') testing were derived by using propagation of errors. Estimates made with the resulting functions and slow crack growth data for a sapphire window were compared to the results of Monte Carlo simulations. The functions for estimation of the variances of the parameters were derived both with and without logarithmic transformation of the initial slow crack growth equations. The transformation was performed to make the functions both more linear and more normal. Comparison of the Monte Carlo results and the closed form expressions derived with propagation of errors indicated that linearization is not required for good estimates of the variances of parameters n and D by the propagation of errors method. However, good estimates variances of the parameters B and A(sup *) could only be made when the starting slow crack growth equation was transformed and the coefficients of variation of the input parameters were not too large. This was partially a result of the skewered distributions of B and A(sup *). Parametric variation of the input parameters was used to determine an acceptable range for using closed form approximate equations derived from propagation of errors.
Júnez-Ferreira, H E; Herrera, G S; González-Hita, L; Cardona, A; Mora-Rodríguez, J
2016-01-01
A new method for the optimal design of groundwater quality monitoring networks is introduced in this paper. Various indicator parameters were considered simultaneously and tested for the Irapuato-Valle aquifer in Mexico. The steps followed in the design were (1) establishment of the monitoring network objectives, (2) definition of a groundwater quality conceptual model for the study area, (3) selection of the parameters to be sampled, and (4) selection of a monitoring network by choosing the well positions that minimize the estimate error variance of the selected indicator parameters. Equal weight for each parameter was given to most of the aquifer positions and a higher weight to priority zones. The objective for the monitoring network in the specific application was to obtain a general reconnaissance of the water quality, including water types, water origin, and first indications of contamination. Water quality indicator parameters were chosen in accordance with this objective, and for the selection of the optimal monitoring sites, it was sought to obtain a low-uncertainty estimate of these parameters for the entire aquifer and with more certainty in priority zones. The optimal monitoring network was selected using a combination of geostatistical methods, a Kalman filter and a heuristic optimization method. Results show that when monitoring the 69 locations with higher priority order (the optimal monitoring network), the joint average standard error in the study area for all the groundwater quality parameters was approximately 90 % of the obtained with the 140 available sampling locations (the set of pilot wells). This demonstrates that an optimal design can help to reduce monitoring costs, by avoiding redundancy in data acquisition.
Development of a Sonar Oil Tanker Cargo Measurement System.
1980-08-01
used at any one time is dependent upon the capacity of the stripping system. The operation is conducted to strip at least as fast , if not faster, than...operational mode, measure the average value and variance of the thickness of an oil layer on an intermit - tent sampling basis. Laboratory testing
Sex Differences in Arithmetical Performance Scores: Central Tendency and Variability
ERIC Educational Resources Information Center
Martens, R.; Hurks, P. P. M.; Meijs, C.; Wassenberg, R.; Jolles, J.
2011-01-01
The present study aimed to analyze sex differences in arithmetical performance in a large-scale sample of 390 children (193 boys) frequenting grades 1-9. Past research in this field has focused primarily on average performance, implicitly assuming homogeneity of variance, for which support is scarce. This article examined sex differences in…
Error simulation of paired-comparison-based scaling methods
NASA Astrophysics Data System (ADS)
Cui, Chengwu
2000-12-01
Subjective image quality measurement usually resorts to psycho physical scaling. However, it is difficult to evaluate the inherent precision of these scaling methods. Without knowing the potential errors of the measurement, subsequent use of the data can be misleading. In this paper, the errors on scaled values derived form paired comparison based scaling methods are simulated with randomly introduced proportion of choice errors that follow the binomial distribution. Simulation results are given for various combinations of the number of stimuli and the sampling size. The errors are presented in the form of average standard deviation of the scaled values and can be fitted reasonably well with an empirical equation that can be sued for scaling error estimation and measurement design. The simulation proves paired comparison based scaling methods can have large errors on the derived scaled values when the sampling size and the number of stimuli are small. Examples are also given to show the potential errors on actually scaled values of color image prints as measured by the method of paired comparison.
Precision of Four Acoustic Bone Measurement Devices
NASA Technical Reports Server (NTRS)
Miller, Christopher; Feiveson, Alan H.; Shackelford, Linda; Rianon, Nahida; LeBlanc, Adrian
2000-01-01
Though many studies have quantified the precision of various acoustic bone measurement devices, it is difficult to directly compare the results among the studies, because they used disparate subject pools, did not specify the estimation methodology, or did not use consistent definitions for various precision characteristics. In this study, we used a repeated measures design protocol to directly determine the precision characteristics of four acoustic bone measurement devices: the Mechanical Response Tissue Analyzer (MRTA), the UBA-575+, the SoundScan 2000 (S2000), and the Sahara Ultrasound Done Analyzer. Ten men and ten women were scanned on all four devices by two different operators at five discrete time points: Week 1, Week 2, Week 3, Month 3 and Month 6. The percent coefficient of variation (%CV) and standardized coefficient of variation were computed for the following precision characteristics: interoperator effect, operator-subject interaction, short-term error variance, and long-term drift, The MRTA had high interoperator errors for its ulnar and tibial stiffness measures and a large long-term drift in its tibial stiffness measurement. The UBA-575+ exhibited large short-term error variances and long-term drift for all three of its measurements. The S2000's tibial speed of sound measurement showed a high short-term error variance and a significant operator-subject interaction but very good values ( < 1%) for the other precision characteristics. The Sahara seemed to have the best overall performance, but was hampered by a large %CV for short-term error variance in its broadband ultrasound attenuation measure.
Precision of Four Acoustic Bone Measurement Devices
NASA Technical Reports Server (NTRS)
Miller, Christopher; Rianon, Nahid; Feiveson, Alan; Shackelford, Linda; LeBlanc, Adrian
2000-01-01
Though many studies have quantified the precision of various acoustic bone measurement devices, it is difficult to directly compare the results among the studies, because they used disparate subject pools, did not specify the estimation methodology, or did not use consistent definitions for various precision characteristics. In this study, we used a repeated measures design protocol to directly determine the precision characteristics of four acoustic bone measurement devices: the Mechanical Response Tissue Analyzer (MRTA), the UBA-575+, the SoundScan 2000 (S2000), and the Sahara Ultrasound Bone Analyzer. Ten men and ten women were scanned on all four devices by two different operators at five discrete time points: Week 1, Week 2, Week 3, Month 3 and Month 6. The percent coefficient of variation (%CV) and standardized coefficient of variation were computed for the following precision characteristics: interoperator effect, operator-subject interaction, short-term error variance, and long-term drift. The MRTA had high interoperator errors for its ulnar and tibial stiffness measures and a large long-term drift in its tibial stiffness measurement. The UBA-575+ exhibited large short-term error variances and long-term drift for all three of its measurements. The S2000's tibial speed of sound measurement showed a high short-term error variance and a significant operator-subject interaction but very good values (less than 1%) for the other precision characteristics. The Sahara seemed to have the best overall performance, but was hampered by a large %CV for short-term error variance in its broadband ultrasound attenuation measure.
Estimating individual glomerular volume in the human kidney: clinical perspectives.
Puelles, Victor G; Zimanyi, Monika A; Samuel, Terence; Hughson, Michael D; Douglas-Denton, Rebecca N; Bertram, John F; Armitage, James A
2012-05-01
Measurement of individual glomerular volumes (IGV) has allowed the identification of drivers of glomerular hypertrophy in subjects without overt renal pathology. This study aims to highlight the relevance of IGV measurements with possible clinical implications and determine how many profiles must be measured in order to achieve stable size distribution estimates. We re-analysed 2250 IGV estimates obtained using the disector/Cavalieri method in 41 African and 34 Caucasian Americans. Pooled IGV analysis of mean and variance was conducted. Monte-Carlo (Jackknife) simulations determined the effect of the number of sampled glomeruli on mean IGV. Lin's concordance coefficient (R(C)), coefficient of variation (CV) and coefficient of error (CE) measured reliability. IGV mean and variance increased with overweight and hypertensive status. Superficial glomeruli were significantly smaller than juxtamedullary glomeruli in all subjects (P < 0.01), by race (P < 0.05) and in obese individuals (P < 0.01). Subjects with multiple chronic kidney disease (CKD) comorbidities showed significant increases in IGV mean and variability. Overall, mean IGV was particularly reliable with nine or more sampled glomeruli (R(C) > 0.95, <5% difference in CV and CE). These observations were not affected by a reduced sample size and did not disrupt the inverse linear correlation between mean IGV and estimated total glomerular number. Multiple comorbidities for CKD are associated with increased IGV mean and variance within subjects, including overweight, obesity and hypertension. Zonal selection and the number of sampled glomeruli do not represent drawbacks for future longitudinal biopsy-based studies of glomerular size and distribution.
The heterogeneity statistic I(2) can be biased in small meta-analyses.
von Hippel, Paul T
2015-04-14
Estimated effects vary across studies, partly because of random sampling error and partly because of heterogeneity. In meta-analysis, the fraction of variance that is due to heterogeneity is estimated by the statistic I(2). We calculate the bias of I(2), focusing on the situation where the number of studies in the meta-analysis is small. Small meta-analyses are common; in the Cochrane Library, the median number of studies per meta-analysis is 7 or fewer. We use Mathematica software to calculate the expectation and bias of I(2). I(2) has a substantial bias when the number of studies is small. The bias is positive when the true fraction of heterogeneity is small, but the bias is typically negative when the true fraction of heterogeneity is large. For example, with 7 studies and no true heterogeneity, I(2) will overestimate heterogeneity by an average of 12 percentage points, but with 7 studies and 80 percent true heterogeneity, I(2) can underestimate heterogeneity by an average of 28 percentage points. Biases of 12-28 percentage points are not trivial when one considers that, in the Cochrane Library, the median I(2) estimate is 21 percent. The point estimate I(2) should be interpreted cautiously when a meta-analysis has few studies. In small meta-analyses, confidence intervals should supplement or replace the biased point estimate I(2).
Palmprint Based Multidimensional Fuzzy Vault Scheme
Liu, Hailun; Sun, Dongmei; Xiong, Ke; Qiu, Zhengding
2014-01-01
Fuzzy vault scheme (FVS) is one of the most popular biometric cryptosystems for biometric template protection. However, error correcting code (ECC) proposed in FVS is not appropriate to deal with real-valued biometric intraclass variances. In this paper, we propose a multidimensional fuzzy vault scheme (MDFVS) in which a general subspace error-tolerant mechanism is designed and embedded into FVS to handle intraclass variances. Palmprint is one of the most important biometrics; to protect palmprint templates; a palmprint based MDFVS implementation is also presented. Experimental results show that the proposed scheme not only can deal with intraclass variances effectively but also could maintain the accuracy and meanwhile enhance security. PMID:24892094
DOE Office of Scientific and Technical Information (OSTI.GOV)
Lu, Siyuan; Hwang, Youngdeok; Khabibrakhmanov, Ildar
With increasing penetration of solar and wind energy to the total energy supply mix, the pressing need for accurate energy forecasting has become well-recognized. Here we report the development of a machine-learning based model blending approach for statistically combining multiple meteorological models for improving the accuracy of solar/wind power forecast. Importantly, we demonstrate that in addition to parameters to be predicted (such as solar irradiance and power), including additional atmospheric state parameters which collectively define weather situations as machine learning input provides further enhanced accuracy for the blended result. Functional analysis of variance shows that the error of individual modelmore » has substantial dependence on the weather situation. The machine-learning approach effectively reduces such situation dependent error thus produces more accurate results compared to conventional multi-model ensemble approaches based on simplistic equally or unequally weighted model averaging. Validation over an extended period of time results show over 30% improvement in solar irradiance/power forecast accuracy compared to forecasts based on the best individual model.« less
An impact analysis of forecasting methods and forecasting parameters on bullwhip effect
NASA Astrophysics Data System (ADS)
Silitonga, R. Y. H.; Jelly, N.
2018-04-01
Bullwhip effect is an increase of variance of demand fluctuation from downstream to upstream of supply chain. Forecasting methods and forecasting parameters were recognized as some factors that affect bullwhip phenomena. To study these factors, we can develop simulations. There are several ways to simulate bullwhip effect in previous studies, such as mathematical equation modelling, information control modelling, computer program, and many more. In this study a spreadsheet program named Bullwhip Explorer was used to simulate bullwhip effect. Several scenarios were developed to show the change in bullwhip effect ratio because of the difference in forecasting methods and forecasting parameters. Forecasting methods used were mean demand, moving average, exponential smoothing, demand signalling, and minimum expected mean squared error. Forecasting parameters were moving average period, smoothing parameter, signalling factor, and safety stock factor. It showed that decreasing moving average period, increasing smoothing parameter, increasing signalling factor can create bigger bullwhip effect ratio. Meanwhile, safety stock factor had no impact to bullwhip effect.
Atwood, E.L.
1958-01-01
Response bias errors are studied by comparing questionnaire responses from waterfowl hunters using four large public hunting areas with actual hunting data from these areas during two hunting seasons. To the extent that the data permit, the sources of the error in the responses were studied and the contribution of each type to the total error was measured. Response bias errors, including both prestige and memory bias, were found to be very large as compared to non-response and sampling errors. Good fits were obtained with the seasonal kill distribution of the actual hunting data and the negative binomial distribution and a good fit was obtained with the distribution of total season hunting activity and the semi-logarithmic curve. A comparison of the actual seasonal distributions with the questionnaire response distributions revealed that the prestige and memory bias errors are both positive. The comparisons also revealed the tendency for memory bias errors to occur at digit frequencies divisible by five and for prestige bias errors to occur at frequencies which are multiples of the legal daily bag limit. A graphical adjustment of the response distributions was carried out by developing a smooth curve from those frequency classes not included in the predictable biased frequency classes referred to above. Group averages were used in constructing the curve, as suggested by Ezekiel [1950]. The efficiency of the technique described for reducing response bias errors in hunter questionnaire responses on seasonal waterfowl kill is high in large samples. The graphical method is not as efficient in removing response bias errors in hunter questionnaire responses on seasonal hunting activity where an average of 60 percent was removed.
Reproducibility of the six-minute walking test in chronic heart failure patients.
Pinna, G D; Opasich, C; Mazza, A; Tangenti, A; Maestri, R; Sanarico, M
2000-11-30
The six-minute walking test (WT) is used in trials and clinical practice as an easy tool to evaluate the functional capacity of chronic heart failure (CHF) patients. As WT measurements are highly variable both between and within individuals, this study aims at assessing the contribution of the different sources of variation and estimating the reproducibility of the test. A statistical model describing WT measurements as a function of fixed and random effects is proposed and its parameters estimated. We considered 202 stable CHF patients who performed two baseline WTs separated by a 30 minute rest; 49 of them repeated the two tests 3 months later (follow-up control). They had no changes in therapy or major clinical events. Another 31 subjects performed two baseline tests separated by 24 hours. Collected data were analysed using a mixed model methodology. There was no significant difference between measurements taken 30 minutes and 24 hours apart (p = 0.99). A trend effect of 17 (1.4) m (mean (SE)) was consistently found between duplicate tests (p < 0.001). REML estimates of variance components were: 5189 (674) for subject differences in the error-free value; 1280 (304) for subject differences in spontaneous clinical evolution between baseline and follow-up control, and 266 (23) for the within-subject error. Hence, the standard error of measurement was 16.3 m, namely 4 per cent of the average WT performance (403 m) in this sample. The intraclass correlation coefficient was 0.96. We conclude that WT measurements are characterized by good intrasubject reproducibility and excellent reliability. When follow-up studies > or = 3 months are performed, unpredictable changes in individual walking performance due to spontaneous clinical evolution are to be expected. Their clinical significance, however, is not known. Copyright 2000 John Wiley & Sons, Ltd.
Whose IQ is it?--Assessor bias variance in high-stakes psychological assessment.
McDermott, Paul A; Watkins, Marley W; Rhoad, Anna M
2014-03-01
Assessor bias variance exists for a psychological measure when some appreciable portion of the score variation that is assumed to reflect examinees' individual differences (i.e., the relevant phenomena in most psychological assessments) instead reflects differences among the examiners who perform the assessment. Ordinary test reliability estimates and standard errors of measurement do not inherently encompass assessor bias variance. This article reports on the application of multilevel linear modeling to examine the presence and extent of assessor bias in the administration of the Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV) for a sample of 2,783 children evaluated by 448 regional school psychologists for high-stakes special education classification purposes. It was found that nearly all WISC-IV scores conveyed significant and nontrivial amounts of variation that had nothing to do with children's actual individual differences and that the Full Scale IQ and Verbal Comprehension Index scores evidenced quite substantial assessor bias. Implications are explored. 2014 APA
Estimating the extreme low-temperature event using nonparametric methods
NASA Astrophysics Data System (ADS)
D'Silva, Anisha
This thesis presents a new method of estimating the one-in-N low temperature threshold using a non-parametric statistical method called kernel density estimation applied to daily average wind-adjusted temperatures. We apply our One-in-N Algorithm to local gas distribution companies (LDCs), as they have to forecast the daily natural gas needs of their consumers. In winter, demand for natural gas is high. Extreme low temperature events are not directly related to an LDCs gas demand forecasting, but knowledge of extreme low temperatures is important to ensure that an LDC has enough capacity to meet customer demands when extreme low temperatures are experienced. We present a detailed explanation of our One-in-N Algorithm and compare it to the methods using the generalized extreme value distribution, the normal distribution, and the variance-weighted composite distribution. We show that our One-in-N Algorithm estimates the one-in- N low temperature threshold more accurately than the methods using the generalized extreme value distribution, the normal distribution, and the variance-weighted composite distribution according to root mean square error (RMSE) measure at a 5% level of significance. The One-in- N Algorithm is tested by counting the number of times the daily average wind-adjusted temperature is less than or equal to the one-in- N low temperature threshold.
Comparison of Grouping Schemes for Exposure to Total Dust in Cement Factories in Korea.
Koh, Dong-Hee; Kim, Tae-Woo; Jang, Seung Hee; Ryu, Hyang-Woo; Park, Donguk
2015-08-01
The purpose of this study was to evaluate grouping schemes for exposure to total dust in cement industry workers using non-repeated measurement data. In total, 2370 total dust measurements taken from nine Portland cement factories in 1995-2009 were analyzed. Various grouping schemes were generated based on work process, job, factory, or average exposure. To characterize variance components of each grouping scheme, we developed mixed-effects models with a B-spline time trend incorporated as fixed effects and a grouping variable incorporated as a random effect. Using the estimated variance components, elasticity was calculated. To compare the prediction performances of different grouping schemes, 10-fold cross-validation tests were conducted, and root mean squared errors and pooled correlation coefficients were calculated for each grouping scheme. The five exposure groups created a posteriori by ranking job and factory combinations according to average dust exposure showed the best prediction performance and highest elasticity among various grouping schemes. Our findings suggest a grouping method based on ranking of job, and factory combinations would be the optimal choice in this population. Our grouping method may aid exposure assessment efforts in similar occupational settings, minimizing the misclassification of exposures. © The Author 2015. Published by Oxford University Press on behalf of the British Occupational Hygiene Society.
Contemporary skull development - palatal angle analysis.
Dostalova, T; Eliasova, H; Gabcova, D; Feberova, J; Kaminek, M
2015-01-01
The palatal angle is an important angle of the craniofacial complex. It is significant for the diagnosis of craniofacial disorders mainly for nasopharyngeal soft-tissue patterns.Background The dentists and otorhinolaryngologists use this relationship to establish proper treatment mechanics and evaluate facial profile. The aims of this study were to provide comparative cephalometric analyses of historical and contemporary skulls. A total of 190 cephalograms of 2 groups of subjects were evaluated. Dolphin Imaging 11.0 - Cephalometric Tracing Analysis was used for the analysis. Unpaired two-tailed t-test assuming equality of variances was used for all variables (at the significance level p = 0.0001). The -modern forensic skulls had larger palatal angle at average value of 8.60 degrees ± 4.35, than that of archeological ones, the average value of which was 6.50 degrees ± 3.92. The difference was found significant. Unpaired two-tailed t-test assuming equality of variances showed that historical and contemporary skulls had statistically significant results. The difference was -2.09 with standard error of 0.60 (95% confidence interval from -3.29 to -0.89). Two-tailed probability attained value of P was less than 0.0001. The difference between both groups was found significant. An increase in the palatal angle can be directly connected with anterior rotation of upper jaw(Tab. 2, Fig. 5, Ref. 19).
Accounting for Relatedness in Family Based Genetic Association Studies
McArdle, P.F.; O’Connell, J.R.; Pollin, T.I.; Baumgarten, M.; Shuldiner, A.R.; Peyser, P.A.; Mitchell, B.D.
2007-01-01
Objective Assess the differences in point estimates, power and type 1 error rates when accounting for and ignoring family structure in genetic tests of association. Methods We compare by simulation the performance of analytic models using variance components to account for family structure and regression models that ignore relatedness for a range of possible family based study designs (i.e., sib pairs vs. large sibships vs. nuclear families vs. extended families). Results Our analyses indicate that effect size estimates and power are not significantly affected by ignoring family structure. Type 1 error rates increase when family structure is ignored, as density of family structures increases, and as trait heritability increases. For discrete traits with moderate levels of heritability and across many common sampling designs, type 1 error rates rise from a nominal 0.05 to 0.11. Conclusion Ignoring family structure may be useful in screening although it comes at a cost of a increased type 1 error rate, the magnitude of which depends on trait heritability and pedigree configuration. PMID:17570925
NASA Technical Reports Server (NTRS)
Schutz, Bob E.; Baker, Gregory A.
1997-01-01
The recovery of a high resolution geopotential from satellite gradiometer observations motivates the examination of high performance computational techniques. The primary subject matter addresses specifically the use of satellite gradiometer and GPS observations to form and invert the normal matrix associated with a large degree and order geopotential solution. Memory resident and out-of-core parallel linear algebra techniques along with data parallel batch algorithms form the foundation of the least squares application structure. A secondary topic includes the adoption of object oriented programming techniques to enhance modularity and reusability of code. Applications implementing the parallel and object oriented methods successfully calculate the degree variance for a degree and order 110 geopotential solution on 32 processors of the Cray T3E. The memory resident gradiometer application exhibits an overall application performance of 5.4 Gflops, and the out-of-core linear solver exhibits an overall performance of 2.4 Gflops. The combination solution derived from a sun synchronous gradiometer orbit produce average geoid height variances of 17 millimeters.
NASA Astrophysics Data System (ADS)
Baker, Gregory Allen
The recovery of a high resolution geopotential from satellite gradiometer observations motivates the examination of high performance computational techniques. The primary subject matter addresses specifically the use of satellite gradiometer and GPS observations to form and invert the normal matrix associated with a large degree and order geopotential solution. Memory resident and out-of-core parallel linear algebra techniques along with data parallel batch algorithms form the foundation of the least squares application structure. A secondary topic includes the adoption of object oriented programming techniques to enhance modularity and reusability of code. Applications implementing the parallel and object oriented methods successfully calculate the degree variance for a degree and order 110 geopotential solution on 32 processors of the Cray T3E. The memory resident gradiometer application exhibits an overall application performance of 5.4 Gflops, and the out-of-core linear solver exhibits an overall performance of 2.4 Gflops. The combination solution derived from a sun synchronous gradiometer orbit produce average geoid height variances of 17 millimeters.
Candela, L.; Olea, R.A.; Custodio, E.
1988-01-01
Groundwater quality observation networks are examples of discontinuous sampling on variables presenting spatial continuity and highly skewed frequency distributions. Anywhere in the aquifer, lognormal kriging provides estimates of the variable being sampled and a standard error of the estimate. The average and the maximum standard error within the network can be used to dynamically improve the network sampling efficiency or find a design able to assure a given reliability level. The approach does not require the formulation of any physical model for the aquifer or any actual sampling of hypothetical configurations. A case study is presented using the network monitoring salty water intrusion into the Llobregat delta confined aquifer, Barcelona, Spain. The variable chloride concentration used to trace the intrusion exhibits sudden changes within short distances which make the standard error fairly invariable to changes in sampling pattern and to substantial fluctuations in the number of wells. ?? 1988.
Compounding approach for univariate time series with nonstationary variances
NASA Astrophysics Data System (ADS)
Schäfer, Rudi; Barkhofen, Sonja; Guhr, Thomas; Stöckmann, Hans-Jürgen; Kuhl, Ulrich
2015-12-01
A defining feature of nonstationary systems is the time dependence of their statistical parameters. Measured time series may exhibit Gaussian statistics on short time horizons, due to the central limit theorem. The sample statistics for long time horizons, however, averages over the time-dependent variances. To model the long-term statistical behavior, we compound the local distribution with the distribution of its parameters. Here, we consider two concrete, but diverse, examples of such nonstationary systems: the turbulent air flow of a fan and a time series of foreign exchange rates. Our main focus is to empirically determine the appropriate parameter distribution for the compounding approach. To this end, we extract the relevant time scales by decomposing the time signals into windows and determine the distribution function of the thus obtained local variances.
Compounding approach for univariate time series with nonstationary variances.
Schäfer, Rudi; Barkhofen, Sonja; Guhr, Thomas; Stöckmann, Hans-Jürgen; Kuhl, Ulrich
2015-12-01
A defining feature of nonstationary systems is the time dependence of their statistical parameters. Measured time series may exhibit Gaussian statistics on short time horizons, due to the central limit theorem. The sample statistics for long time horizons, however, averages over the time-dependent variances. To model the long-term statistical behavior, we compound the local distribution with the distribution of its parameters. Here, we consider two concrete, but diverse, examples of such nonstationary systems: the turbulent air flow of a fan and a time series of foreign exchange rates. Our main focus is to empirically determine the appropriate parameter distribution for the compounding approach. To this end, we extract the relevant time scales by decomposing the time signals into windows and determine the distribution function of the thus obtained local variances.
Evaluation of causes and frequency of medication errors during information technology downtime.
Hanuscak, Tara L; Szeinbach, Sheryl L; Seoane-Vazquez, Enrique; Reichert, Brendan J; McCluskey, Charles F
2009-06-15
The causes and frequency of medication errors occurring during information technology downtime were evaluated. Individuals from a convenience sample of 78 hospitals who were directly responsible for supporting and maintaining clinical information systems (CISs) and automated dispensing systems (ADSs) were surveyed using an online tool between February 2007 and May 2007 to determine if medication errors were reported during periods of system downtime. The errors were classified using the National Coordinating Council for Medication Error Reporting and Prevention severity scoring index. The percentage of respondents reporting downtime was estimated. Of the 78 eligible hospitals, 32 respondents with CIS and ADS responsibilities completed the online survey for a response rate of 41%. For computerized prescriber order entry, patch installations and system upgrades caused an average downtime of 57% over a 12-month period. Lost interface and interface malfunction were reported for centralized and decentralized ADSs, with an average downtime response of 34% and 29%, respectively. The average downtime response was 31% for software malfunctions linked to clinical decision-support systems. Although patient harm did not result from 30 (54%) medication errors, the potential for harm was present for 9 (16%) of these errors. Medication errors occurred during CIS and ADS downtime despite the availability of backup systems and standard protocols to handle periods of system downtime. Efforts should be directed to reduce the frequency and length of down-time in order to minimize medication errors during such downtime.
NASA Technical Reports Server (NTRS)
Craig, R. G. (Principal Investigator)
1983-01-01
Richmond, Virginia and Denver, Colorado were study sites in an effort to determine the effect of autocorrelation on the accuracy of a parallelopiped classifier of LANDSAT digital data. The autocorrelation was assumed to decay to insignificant levels when sampled at distances of at least ten pixels. Spectral themes developed using blocks of adjacent pixels, and using groups of pixels spaced at least 10 pixels apart were used. Effects of geometric distortions were minimized by using only pixels from the interiors of land cover sections. Accuracy was evaluated for three classes; agriculture, residential and "all other"; both type 1 and type 2 errors were evaluated by means of overall classification accuracy. All classes give comparable results. Accuracy is approximately the same in both techniques; however, the variance in accuracy is significantly higher using the themes developed from autocorrelated data. The vectors of mean spectral response were nearly identical regardless of sampling method used. The estimated variances were much larger when using autocorrelated pixels.
NASA Astrophysics Data System (ADS)
Hernández, Mario R.; Francés, Félix
2015-04-01
One phase of the hydrological models implementation process, significantly contributing to the hydrological predictions uncertainty, is the calibration phase in which values of the unknown model parameters are tuned by optimizing an objective function. An unsuitable error model (e.g. Standard Least Squares or SLS) introduces noise into the estimation of the parameters. The main sources of this noise are the input errors and the hydrological model structural deficiencies. Thus, the biased calibrated parameters cause the divergence model phenomenon, where the errors variance of the (spatially and temporally) forecasted flows far exceeds the errors variance in the fitting period, and provoke the loss of part or all of the physical meaning of the modeled processes. In other words, yielding a calibrated hydrological model which works well, but not for the right reasons. Besides, an unsuitable error model yields a non-reliable predictive uncertainty assessment. Hence, with the aim of prevent all these undesirable effects, this research focuses on the Bayesian joint inference (BJI) of both the hydrological and error model parameters, considering a general additive (GA) error model that allows for correlation, non-stationarity (in variance and bias) and non-normality of model residuals. As hydrological model, it has been used a conceptual distributed model called TETIS, with a particular split structure of the effective model parameters. Bayesian inference has been performed with the aid of a Markov Chain Monte Carlo (MCMC) algorithm called Dream-ZS. MCMC algorithm quantifies the uncertainty of the hydrological and error model parameters by getting the joint posterior probability distribution, conditioned on the observed flows. The BJI methodology is a very powerful and reliable tool, but it must be used correctly this is, if non-stationarity in errors variance and bias is modeled, the Total Laws must be taken into account. The results of this research show that the application of BJI with a GA error model outperforms the hydrological parameters robustness (diminishing the divergence model phenomenon) and improves the reliability of the streamflow predictive distribution, in respect of the results of a bad error model as SLS. Finally, the most likely prediction in a validation period, for both BJI+GA and SLS error models shows a similar performance.
Jackknife variance of the partial area under the empirical receiver operating characteristic curve.
Bandos, Andriy I; Guo, Ben; Gur, David
2017-04-01
Receiver operating characteristic analysis provides an important methodology for assessing traditional (e.g., imaging technologies and clinical practices) and new (e.g., genomic studies, biomarker development) diagnostic problems. The area under the clinically/practically relevant part of the receiver operating characteristic curve (partial area or partial area under the receiver operating characteristic curve) is an important performance index summarizing diagnostic accuracy at multiple operating points (decision thresholds) that are relevant to actual clinical practice. A robust estimate of the partial area under the receiver operating characteristic curve is provided by the area under the corresponding part of the empirical receiver operating characteristic curve. We derive a closed-form expression for the jackknife variance of the partial area under the empirical receiver operating characteristic curve. Using the derived analytical expression, we investigate the differences between the jackknife variance and a conventional variance estimator. The relative properties in finite samples are demonstrated in a simulation study. The developed formula enables an easy way to estimate the variance of the empirical partial area under the receiver operating characteristic curve, thereby substantially reducing the computation burden, and provides important insight into the structure of the variability. We demonstrate that when compared with the conventional approach, the jackknife variance has substantially smaller bias, and leads to a more appropriate type I error rate of the Wald-type test. The use of the jackknife variance is illustrated in the analysis of a data set from a diagnostic imaging study.
Measuring kinetics of complex single ion channel data using mean-variance histograms.
Patlak, J B
1993-07-01
The measurement of single ion channel kinetics is difficult when those channels exhibit subconductance events. When the kinetics are fast, and when the current magnitudes are small, as is the case for Na+, Ca2+, and some K+ channels, these difficulties can lead to serious errors in the estimation of channel kinetics. I present here a method, based on the construction and analysis of mean-variance histograms, that can overcome these problems. A mean-variance histogram is constructed by calculating the mean current and the current variance within a brief "window" (a set of N consecutive data samples) superimposed on the digitized raw channel data. Systematic movement of this window over the data produces large numbers of mean-variance pairs which can be assembled into a two-dimensional histogram. Defined current levels (open, closed, or sublevel) appear in such plots as low variance regions. The total number of events in such low variance regions is estimated by curve fitting and plotted as a function of window width. This function decreases with the same time constants as the original dwell time probability distribution for each of the regions. The method can therefore be used: 1) to present a qualitative summary of the single channel data from which the signal-to-noise ratio, open channel noise, steadiness of the baseline, and number of conductance levels can be quickly determined; 2) to quantify the dwell time distribution in each of the levels exhibited. In this paper I present the analysis of a Na+ channel recording that had a number of complexities. The signal-to-noise ratio was only about 8 for the main open state, open channel noise, and fast flickers to other states were present, as were a substantial number of subconductance states. "Standard" half-amplitude threshold analysis of these data produce open and closed time histograms that were well fitted by the sum of two exponentials, but with apparently erroneous time constants, whereas the mean-variance histogram technique provided a more credible analysis of the open, closed, and subconductance times for the patch. I also show that the method produces accurate results on simulated data in a wide variety of conditions, whereas the half-amplitude method, when applied to complex simulated data shows the same errors as were apparent in the real data. The utility and the limitations of this new method are discussed.
Multisite Reliability of Cognitive BOLD Data
Brown, Gregory G.; Mathalon, Daniel H.; Stern, Hal; Ford, Judith; Mueller, Bryon; Greve, Douglas N.; McCarthy, Gregory; Voyvodic, Jim; Glover, Gary; Diaz, Michele; Yetter, Elizabeth; Burak Ozyurt, I.; Jorgensen, Kasper W.; Wible, Cynthia G.; Turner, Jessica A.; Thompson, Wesley K.; Potkin, Steven G.
2010-01-01
Investigators perform multi-site functional magnetic resonance imaging studies to increase statistical power, to enhance generalizability, and to improve the likelihood of sampling relevant subgroups. Yet undesired site variation in imaging methods could off-set these potential advantages. We used variance components analysis to investigate sources of variation in the blood oxygen level dependent (BOLD) signal across four 3T magnets in voxelwise and region of interest (ROI) analyses. Eighteen participants traveled to four magnet sites to complete eight runs of a working memory task involving emotional or neutral distraction. Person variance was more than 10 times larger than site variance for five of six ROIs studied. Person-by-site interactions, however, contributed sizable unwanted variance to the total. Averaging over runs increased between-site reliability, with many voxels showing good to excellent between-site reliability when eight runs were averaged and regions of interest showing fair to good reliability. Between-site reliability depended on the specific functional contrast analyzed in addition to the number of runs averaged. Although median effect size was correlated with between-site reliability, dissociations were observed for many voxels. Brain regions where the pooled effect size was large but between-site reliability was poor were associated with reduced individual differences. Brain regions where the pooled effect size was small but between-site reliability was excellent were associated with a balance of participants who displayed consistently positive or consistently negative BOLD responses. Although between-site reliability of BOLD data can be good to excellent, acquiring highly reliable data requires robust activation paradigms, ongoing quality assurance, and careful experimental control. PMID:20932915
Deflation as a method of variance reduction for estimating the trace of a matrix inverse
DOE Office of Scientific and Technical Information (OSTI.GOV)
Gambhir, Arjun Singh; Stathopoulos, Andreas; Orginos, Kostas
Many fields require computing the trace of the inverse of a large, sparse matrix. The typical method used for such computations is the Hutchinson method which is a Monte Carlo (MC) averaging over matrix quadratures. To improve its convergence, several variance reductions techniques have been proposed. In this paper, we study the effects of deflating the near null singular value space. We make two main contributions. First, we analyze the variance of the Hutchinson method as a function of the deflated singular values and vectors. Although this provides good intuition in general, by assuming additionally that the singular vectors aremore » random unitary matrices, we arrive at concise formulas for the deflated variance that include only the variance and mean of the singular values. We make the remarkable observation that deflation may increase variance for Hermitian matrices but not for non-Hermitian ones. This is a rare, if not unique, property where non-Hermitian matrices outperform Hermitian ones. The theory can be used as a model for predicting the benefits of deflation. Second, we use deflation in the context of a large scale application of "disconnected diagrams" in Lattice QCD. On lattices, Hierarchical Probing (HP) has previously provided an order of magnitude of variance reduction over MC by removing "error" from neighboring nodes of increasing distance in the lattice. Although deflation used directly on MC yields a limited improvement of 30% in our problem, when combined with HP they reduce variance by a factor of over 150 compared to MC. For this, we pre-computated 1000 smallest singular values of an ill-conditioned matrix of size 25 million. Furthermore, using PRIMME and a domain-specific Algebraic Multigrid preconditioner, we perform one of the largest eigenvalue computations in Lattice QCD at a fraction of the cost of our trace computation.« less
Deflation as a method of variance reduction for estimating the trace of a matrix inverse
Gambhir, Arjun Singh; Stathopoulos, Andreas; Orginos, Kostas
2017-04-06
Many fields require computing the trace of the inverse of a large, sparse matrix. The typical method used for such computations is the Hutchinson method which is a Monte Carlo (MC) averaging over matrix quadratures. To improve its convergence, several variance reductions techniques have been proposed. In this paper, we study the effects of deflating the near null singular value space. We make two main contributions. First, we analyze the variance of the Hutchinson method as a function of the deflated singular values and vectors. Although this provides good intuition in general, by assuming additionally that the singular vectors aremore » random unitary matrices, we arrive at concise formulas for the deflated variance that include only the variance and mean of the singular values. We make the remarkable observation that deflation may increase variance for Hermitian matrices but not for non-Hermitian ones. This is a rare, if not unique, property where non-Hermitian matrices outperform Hermitian ones. The theory can be used as a model for predicting the benefits of deflation. Second, we use deflation in the context of a large scale application of "disconnected diagrams" in Lattice QCD. On lattices, Hierarchical Probing (HP) has previously provided an order of magnitude of variance reduction over MC by removing "error" from neighboring nodes of increasing distance in the lattice. Although deflation used directly on MC yields a limited improvement of 30% in our problem, when combined with HP they reduce variance by a factor of over 150 compared to MC. For this, we pre-computated 1000 smallest singular values of an ill-conditioned matrix of size 25 million. Furthermore, using PRIMME and a domain-specific Algebraic Multigrid preconditioner, we perform one of the largest eigenvalue computations in Lattice QCD at a fraction of the cost of our trace computation.« less
Foster, Guy M.
2014-01-01
The Neosho River and its primary tributary, the Cottonwood River, are the primary sources of inflow to the John Redmond Reservoir in east-central Kansas. Sedimentation rate in the John Redmond Reservoir was estimated as 743 acre-feet per year for 1964–2006. This estimated sedimentation rate is more than 80 percent larger than the projected design sedimentation rate of 404 acre-feet per year, and resulted in a loss of 40 percent of the conservation pool since its construction in 1964. To reduce sediment input into the reservoir, the Kansas Water Office implemented stream bank stabilization techniques along an 8.3 mile reach of the Neosho River during 2010 through 2011. The U.S. Geological Survey, in cooperation with the Kansas Water Office and funded in part through the Kansas State Water Plan Fund, operated continuous real-time water-quality monitors upstream and downstream from stream bank stabilization efforts before, during, and after construction. Continuously measured water-quality properties include streamflow, specific conductance, water temperature, and turbidity. Discrete sediment samples were collected from June 2009 through September 2012 and analyzed for suspended-sediment concentration (SSC), percentage of sediments less than 63 micrometers (sand-fine break), and loss of material on ignition (analogous to amount of organic matter). Regression models were developed to establish relations between discretely measured SSC samples, and turbidity or streamflow to estimate continuously SSC. Continuous water-quality monitors represented between 96 and 99 percent of the cross-sectional variability for turbidity, and had slopes between 0.91 and 0.98. Because consistent bias was not observed, values from continuous water-quality monitors were considered representative of stream conditions. On average, turbidity-based SSC models explained 96 percent of the variance in SSC. Streamflow-based regressions explained 53 to 60 percent of the variance. Mean squared prediction error for turbidity-based regression relations ranged from -32 to 48 percent, whereas mean square prediction error for streamflow-based regressions ranged from -69 to 218 percent. These models are useful for evaluating the variability of SSC during rapidly changing conditions, computing loads and yields to assess SSC transport through the watershed, and for providing more accurate load estimates compared to streamflow-only based estimation methods used in the past. These models can be used to evaluate the efficacy of streambank stabilization efforts.
Hand-writing motion tracking with vision-inertial sensor fusion: calibration and error correction.
Zhou, Shengli; Fei, Fei; Zhang, Guanglie; Liu, Yunhui; Li, Wen J
2014-08-25
The purpose of this study was to improve the accuracy of real-time ego-motion tracking through inertial sensor and vision sensor fusion. Due to low sampling rates supported by web-based vision sensor and accumulation of errors in inertial sensors, ego-motion tracking with vision sensors is commonly afflicted by slow updating rates, while motion tracking with inertial sensor suffers from rapid deterioration in accuracy with time. This paper starts with a discussion of developed algorithms for calibrating two relative rotations of the system using only one reference image. Next, stochastic noises associated with the inertial sensor are identified using Allan Variance analysis, and modeled according to their characteristics. Finally, the proposed models are incorporated into an extended Kalman filter for inertial sensor and vision sensor fusion. Compared with results from conventional sensor fusion models, we have shown that ego-motion tracking can be greatly enhanced using the proposed error correction model.
Xiao, Yongling; Abrahamowicz, Michal
2010-03-30
We propose two bootstrap-based methods to correct the standard errors (SEs) from Cox's model for within-cluster correlation of right-censored event times. The cluster-bootstrap method resamples, with replacement, only the clusters, whereas the two-step bootstrap method resamples (i) the clusters, and (ii) individuals within each selected cluster, with replacement. In simulations, we evaluate both methods and compare them with the existing robust variance estimator and the shared gamma frailty model, which are available in statistical software packages. We simulate clustered event time data, with latent cluster-level random effects, which are ignored in the conventional Cox's model. For cluster-level covariates, both proposed bootstrap methods yield accurate SEs, and type I error rates, and acceptable coverage rates, regardless of the true random effects distribution, and avoid serious variance under-estimation by conventional Cox-based standard errors. However, the two-step bootstrap method over-estimates the variance for individual-level covariates. We also apply the proposed bootstrap methods to obtain confidence bands around flexible estimates of time-dependent effects in a real-life analysis of cluster event times.
Improving lidar turbulence estimates for wind energy
NASA Astrophysics Data System (ADS)
Newman, J. F.; Clifton, A.; Churchfield, M. J.; Klein, P.
2016-09-01
Remote sensing devices (e.g., lidars) are quickly becoming a cost-effective and reliable alternative to meteorological towers for wind energy applications. Although lidars can measure mean wind speeds accurately, these devices measure different values of turbulence intensity (TI) than an instrument on a tower. In response to these issues, a lidar TI error reduction model was recently developed for commercially available lidars. The TI error model first applies physics-based corrections to the lidar measurements, then uses machine-learning techniques to further reduce errors in lidar TI estimates. The model was tested at two sites in the Southern Plains where vertically profiling lidars were collocated with meteorological towers. Results indicate that the model works well under stable conditions but cannot fully mitigate the effects of variance contamination under unstable conditions. To understand how variance contamination affects lidar TI estimates, a new set of equations was derived in previous work to characterize the actual variance measured by a lidar. Terms in these equations were quantified using a lidar simulator and modeled wind field, and the new equations were then implemented into the TI error model.
Improving Lidar Turbulence Estimates for Wind Energy: Preprint
DOE Office of Scientific and Technical Information (OSTI.GOV)
Newman, Jennifer; Clifton, Andrew; Churchfield, Matthew
2016-10-01
Remote sensing devices (e.g., lidars) are quickly becoming a cost-effective and reliable alternative to meteorological towers for wind energy applications. Although lidars can measure mean wind speeds accurately, these devices measure different values of turbulence intensity (TI) than an instrument on a tower. In response to these issues, a lidar TI error reduction model was recently developed for commercially available lidars. The TI error model first applies physics-based corrections to the lidar measurements, then uses machine-learning techniques to further reduce errors in lidar TI estimates. The model was tested at two sites in the Southern Plains where vertically profiling lidarsmore » were collocated with meteorological towers. Results indicate that the model works well under stable conditions but cannot fully mitigate the effects of variance contamination under unstable conditions. To understand how variance contamination affects lidar TI estimates, a new set of equations was derived in previous work to characterize the actual variance measured by a lidar. Terms in these equations were quantified using a lidar simulator and modeled wind field, and the new equations were then implemented into the TI error model.« less
Improving Lidar Turbulence Estimates for Wind Energy
Newman, Jennifer F.; Clifton, Andrew; Churchfield, Matthew J.; ...
2016-10-03
Remote sensing devices (e.g., lidars) are quickly becoming a cost-effective and reliable alternative to meteorological towers for wind energy applications. Although lidars can measure mean wind speeds accurately, these devices measure different values of turbulence intensity (TI) than an instrument on a tower. In response to these issues, a lidar TI error reduction model was recently developed for commercially available lidars. The TI error model first applies physics-based corrections to the lidar measurements, then uses machine-learning techniques to further reduce errors in lidar TI estimates. The model was tested at two sites in the Southern Plains where vertically profiling lidarsmore » were collocated with meteorological towers. Results indicate that the model works well under stable conditions but cannot fully mitigate the effects of variance contamination under unstable conditions. To understand how variance contamination affects lidar TI estimates, a new set of equations was derived in previous work to characterize the actual variance measured by a lidar. Terms in these equations were quantified using a lidar simulator and modeled wind field, and the new equations were then implemented into the TI error model.« less
Prediction of true test scores from observed item scores and ancillary data.
Haberman, Shelby J; Yao, Lili; Sinharay, Sandip
2015-05-01
In many educational tests which involve constructed responses, a traditional test score is obtained by adding together item scores obtained through holistic scoring by trained human raters. For example, this practice was used until 2008 in the case of GRE(®) General Analytical Writing and until 2009 in the case of TOEFL(®) iBT Writing. With use of natural language processing, it is possible to obtain additional information concerning item responses from computer programs such as e-rater(®). In addition, available information relevant to examinee performance may include scores on related tests. We suggest application of standard results from classical test theory to the available data to obtain best linear predictors of true traditional test scores. In performing such analysis, we require estimation of variances and covariances of measurement errors, a task which can be quite difficult in the case of tests with limited numbers of items and with multiple measurements per item. As a consequence, a new estimation method is suggested based on samples of examinees who have taken an assessment more than once. Such samples are typically not random samples of the general population of examinees, so that we apply statistical adjustment methods to obtain the needed estimated variances and covariances of measurement errors. To examine practical implications of the suggested methods of analysis, applications are made to GRE General Analytical Writing and TOEFL iBT Writing. Results obtained indicate that substantial improvements are possible both in terms of reliability of scoring and in terms of assessment reliability. © 2015 The British Psychological Society.
Semenova, Vera A.; Steward-Clark, Evelene; Maniatis, Panagiotis; Epperson, Monica; Sabnis, Amit; Schiffer, Jarad
2017-01-01
To improve surge testing capability for a response to a release of Bacillus anthracis, the CDC anti-Protective Antigen (PA) IgG Enzyme-Linked Immunosorbent Assay (ELISA) was re-designed into a high throughput screening format. The following assay performance parameters were evaluated: goodness of fit (measured as the mean reference standard r2), accuracy (measured as percent error), precision (measured as coefficient of variance (CV)), lower limit of detection (LLOD), lower limit of quantification (LLOQ), dilutional linearity, diagnostic sensitivity (DSN) and diagnostic specificity (DSP). The paired sets of data for each sample were evaluated by Concordance Correlation Coefficient (CCC) analysis. The goodness of fit was 0.999; percent error between the expected and observed concentration for each sample ranged from −4.6% to 14.4%. The coefficient of variance ranged from 9.0% to 21.2%. The assay LLOQ was 2.6 μg/mL. The regression analysis results for dilutional linearity data were r2 = 0.952, slope = 1.02 and intercept = −0.03. CCC between assays was 0.974 for the median concentration of serum samples. The accuracy and precision components of CCC were 0.997 and 0.977, respectively. This high throughput screening assay is precise, accurate, sensitive and specific. Anti-PA IgG concentrations determined using two different assays proved high levels of agreement. The method will improve surge testing capability 18-fold from 4 to 72 sera per assay plate. PMID:27814939
Semenova, Vera A; Steward-Clark, Evelene; Maniatis, Panagiotis; Epperson, Monica; Sabnis, Amit; Schiffer, Jarad
2017-01-01
To improve surge testing capability for a response to a release of Bacillus anthracis, the CDC anti-Protective Antigen (PA) IgG Enzyme-Linked Immunosorbent Assay (ELISA) was re-designed into a high throughput screening format. The following assay performance parameters were evaluated: goodness of fit (measured as the mean reference standard r 2 ), accuracy (measured as percent error), precision (measured as coefficient of variance (CV)), lower limit of detection (LLOD), lower limit of quantification (LLOQ), dilutional linearity, diagnostic sensitivity (DSN) and diagnostic specificity (DSP). The paired sets of data for each sample were evaluated by Concordance Correlation Coefficient (CCC) analysis. The goodness of fit was 0.999; percent error between the expected and observed concentration for each sample ranged from -4.6% to 14.4%. The coefficient of variance ranged from 9.0% to 21.2%. The assay LLOQ was 2.6 μg/mL. The regression analysis results for dilutional linearity data were r 2 = 0.952, slope = 1.02 and intercept = -0.03. CCC between assays was 0.974 for the median concentration of serum samples. The accuracy and precision components of CCC were 0.997 and 0.977, respectively. This high throughput screening assay is precise, accurate, sensitive and specific. Anti-PA IgG concentrations determined using two different assays proved high levels of agreement. The method will improve surge testing capability 18-fold from 4 to 72 sera per assay plate. Published by Elsevier Ltd.
Model determination in a case of heterogeneity of variance using sampling techniques.
Varona, L; Moreno, C; Garcia-Cortes, L A; Altarriba, J
1997-01-12
A sampling determination procedure has been described in a case of heterogeneity of variance. The procedure makes use of the predictive distributions of each data given the rest of the data and the structure of the assumed model. The computation of these predictive distributions is carried out using a Gibbs Sampling procedure. The final criterion to compare between models is the Mean Square Error between the expectation of predictive distributions and real data. The procedure has been applied to a data set of weight at 210 days in the Spanish Pirenaica beef cattle breed. Three proposed models have been compared: (a) Single Trait Animal Model; (b) Heterogeneous Variance Animal Model; and (c) Multiple Trait Animal Model. After applying the procedure, the most adjusted model was the Heterogeneous Variance Animal Model. This result is probably due to a compromise between the complexity of the model and the amount of available information. The estimated heritabilities under the preferred model have been 0.489 ± 0.076 for males and 0.331 ± 0.082 for females. RESUMEN: Contraste de modelos en un caso de heterogeneidad de varianzas usando métodos de muestreo Se ha descrito un método de contraste de modelos mediante técnicas de muestreo en un caso de heterogeneidad de varianza entre sexos. El procedimiento utiliza las distribucviones predictivas de cada dato, dado el resto de datos y la estructura del modelo. El criterio para coparar modelos es el error cuadrático medio entre la esperanza de las distribuciones predictivas y los datos reales. El procedimiento se ha aplicado en datos de peso a los 210 días en la raza bovina Pirenaica. Se han propuesto tres posibles modelos: (a) Modelo Animal Unicaracter; (b) Modelo Animal con Varianzas Heterogéneas; (c) Modelo Animal Multicaracter. El modelo mejor ajustado fue el Modelo Animal con Varianzas Heterogéneas. Este resultado es probablemente debido a un compromiso entre la complejidad del modelo y la cantidad de datos disponibles. Las heredabilidades estimadas bajo el modelo preferido han sido 0,489 ± 0,076 en los machos y 0,331 ± 0,082 en las hembras. 1997 Blackwell Verlag GmbH.
Online Estimation of Allan Variance Coefficients Based on a Neural-Extended Kalman Filter
Miao, Zhiyong; Shen, Feng; Xu, Dingjie; He, Kunpeng; Tian, Chunmiao
2015-01-01
As a noise analysis method for inertial sensors, the traditional Allan variance method requires the storage of a large amount of data and manual analysis for an Allan variance graph. Although the existing online estimation methods avoid the storage of data and the painful procedure of drawing slope lines for estimation, they require complex transformations and even cause errors during the modeling of dynamic Allan variance. To solve these problems, first, a new state-space model that directly models the stochastic errors to obtain a nonlinear state-space model was established for inertial sensors. Then, a neural-extended Kalman filter algorithm was used to estimate the Allan variance coefficients. The real noises of an ADIS16405 IMU and fiber optic gyro-sensors were analyzed by the proposed method and traditional methods. The experimental results show that the proposed method is more suitable to estimate the Allan variance coefficients than the traditional methods. Moreover, the proposed method effectively avoids the storage of data and can be easily implemented using an online processor. PMID:25625903
NASA Astrophysics Data System (ADS)
Iwema, J.; Rosolem, R.; Baatz, R.; Wagener, T.; Bogena, H. R.
2015-07-01
The Cosmic-Ray Neutron Sensor (CRNS) can provide soil moisture information at scales relevant to hydrometeorological modelling applications. Site-specific calibration is needed to translate CRNS neutron intensities into sensor footprint average soil moisture contents. We investigated temporal sampling strategies for calibration of three CRNS parameterisations (modified N0, HMF, and COSMIC) by assessing the effects of the number of sampling days and soil wetness conditions on the performance of the calibration results while investigating actual neutron intensity measurements, for three sites with distinct climate and land use: a semi-arid site, a temperate grassland, and a temperate forest. When calibrated with 1 year of data, both COSMIC and the modified N0 method performed better than HMF. The performance of COSMIC was remarkably good at the semi-arid site in the USA, while the N0mod performed best at the two temperate sites in Germany. The successful performance of COSMIC at all three sites can be attributed to the benefits of explicitly resolving individual soil layers (which is not accounted for in the other two parameterisations). To better calibrate these parameterisations, we recommend in situ soil sampled to be collected on more than a single day. However, little improvement is observed for sampling on more than 6 days. At the semi-arid site, the N0mod method was calibrated better under site-specific average wetness conditions, whereas HMF and COSMIC were calibrated better under drier conditions. Average soil wetness condition gave better calibration results at the two humid sites. The calibration results for the HMF method were better when calibrated with combinations of days with similar soil wetness conditions, opposed to N0mod and COSMIC, which profited from using days with distinct wetness conditions. Errors in actual neutron intensities were translated to average errors specifically to each site. At the semi-arid site, these errors were below the typical measurement uncertainties from in situ point-scale sensors and satellite remote sensing products. Nevertheless, at the two humid sites, reduction in uncertainty with increasing sampling days only reached typical errors associated with satellite remote sensing products. The outcomes of this study can be used by researchers as a CRNS calibration strategy guideline.
NASA Astrophysics Data System (ADS)
Wilson, M. D.; Durand, M.; Jung, H. C.; Alsdorf, D.
2015-04-01
The Surface Water and Ocean Topography (SWOT) mission, scheduled for launch in 2020, will provide a step-change improvement in the measurement of terrestrial surface-water storage and dynamics. In particular, it will provide the first, routine two-dimensional measurements of water-surface elevations. In this paper, we aimed to (i) characterise and illustrate in two dimensions the errors which may be found in SWOT swath measurements of terrestrial surface water, (ii) simulate the spatio-temporal sampling scheme of SWOT for the Amazon, and (iii) assess the impact of each of these on estimates of water-surface slope and river discharge which may be obtained from SWOT imagery. We based our analysis on a virtual mission for a ~260 km reach of the central Amazon (Solimões) River, using a hydraulic model to provide water-surface elevations according to SWOT spatio-temporal sampling to which errors were added based on a two-dimensional height error spectrum derived from the SWOT design requirements. We thereby obtained water-surface elevation measurements for the Amazon main stem as may be observed by SWOT. Using these measurements, we derived estimates of river slope and discharge and compared them to those obtained directly from the hydraulic model. We found that cross-channel and along-reach averaging of SWOT measurements using reach lengths greater than 4 km for the Solimões and 7.5 km for Purus reduced the effect of systematic height errors, enabling discharge to be reproduced accurately from the water height, assuming known bathymetry and friction. Using cross-sectional averaging and 20 km reach lengths, results show Nash-Sutcliffe model efficiency values of 0.99 for the Solimões and 0.88 for the Purus, with 2.6 and 19.1 % average overall error in discharge, respectively. We extend the results to other rivers worldwide and infer that SWOT-derived discharge estimates may be more accurate for rivers with larger channel widths (permitting a greater level of cross-sectional averaging and the use of shorter reach lengths) and higher water-surface slopes (reducing the proportional impact of slope errors on discharge calculation).
Estimating random errors due to shot noise in backscatter lidar observations.
Liu, Zhaoyan; Hunt, William; Vaughan, Mark; Hostetler, Chris; McGill, Matthew; Powell, Kathleen; Winker, David; Hu, Yongxiang
2006-06-20
We discuss the estimation of random errors due to shot noise in backscatter lidar observations that use either photomultiplier tube (PMT) or avalanche photodiode (APD) detectors. The statistical characteristics of photodetection are reviewed, and photon count distributions of solar background signals and laser backscatter signals are examined using airborne lidar observations at 532 nm using a photon-counting mode APD. Both distributions appear to be Poisson, indicating that the arrival at the photodetector of photons for these signals is a Poisson stochastic process. For Poisson- distributed signals, a proportional, one-to-one relationship is known to exist between the mean of a distribution and its variance. Although the multiplied photocurrent no longer follows a strict Poisson distribution in analog-mode APD and PMT detectors, the proportionality still exists between the mean and the variance of the multiplied photocurrent. We make use of this relationship by introducing the noise scale factor (NSF), which quantifies the constant of proportionality that exists between the root mean square of the random noise in a measurement and the square root of the mean signal. Using the NSF to estimate random errors in lidar measurements due to shot noise provides a significant advantage over the conventional error estimation techniques, in that with the NSF, uncertainties can be reliably calculated from or for a single data sample. Methods for evaluating the NSF are presented. Algorithms to compute the NSF are developed for the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations lidar and tested using data from the Lidar In-space Technology Experiment.
Estimating Random Errors Due to Shot Noise in Backscatter Lidar Observations
NASA Technical Reports Server (NTRS)
Liu, Zhaoyan; Hunt, William; Vaughan, Mark A.; Hostetler, Chris A.; McGill, Matthew J.; Powell, Kathy; Winker, David M.; Hu, Yongxiang
2006-01-01
In this paper, we discuss the estimation of random errors due to shot noise in backscatter lidar observations that use either photomultiplier tube (PMT) or avalanche photodiode (APD) detectors. The statistical characteristics of photodetection are reviewed, and photon count distributions of solar background signals and laser backscatter signals are examined using airborne lidar observations at 532 nm using a photon-counting mode APD. Both distributions appear to be Poisson, indicating that the arrival at the photodetector of photons for these signals is a Poisson stochastic process. For Poisson-distributed signals, a proportional, one-to-one relationship is known to exist between the mean of a distribution and its variance. Although the multiplied photocurrent no longer follows a strict Poisson distribution in analog-mode APD and PMT detectors, the proportionality still exists between the mean and the variance of the multiplied photocurrent. We make use of this relationship by introducing the noise scale factor (NSF), which quantifies the constant of proportionality that exists between the root-mean-square of the random noise in a measurement and the square root of the mean signal. Using the NSF to estimate random errors in lidar measurements due to shot noise provides a significant advantage over the conventional error estimation techniques, in that with the NSF uncertainties can be reliably calculated from/for a single data sample. Methods for evaluating the NSF are presented. Algorithms to compute the NSF are developed for the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) lidar and tested using data from the Lidar In-space Technology Experiment (LITE). OCIS Codes:
Campos, Juliana Alvares Duarte Bonini; Spexoto, Maria Cláudia Bernardes; Serrano, Sergio Vicente; Maroco, João
2016-01-13
The psychometric properties of an instrument should be evaluated routinely when using different samples. This study evaluated the psychometric properties of the Functional Assessment of Cancer Therapy-General (FACT-G) when applied to a sample of Brazilian cancer patients. The face, content, and construct (factorial, convergent, and discriminant) validities of the FACT-G were estimated. Confirmatory factor analysis (CFA) was conducted the ratio chi-square by degrees of freedom (χ (2)/df), the comparative fit index (CFI), the Tucker-Lewis index (TLI), and the root mean square error of approximation (RMSEA) as indices. The invariance of the best model was assessed with multi-group analysis using the difference of chi-squares method (Δχ(2)). Convergent validity was assessed using Average Variance Extracted (AVE) and discriminant validity was determined via correlational analysis. Internal consistency was assessed using the Cronbach's alpha (α) coefficient, and the Composite Reliability (CR) was estimated. A total of 975 cancer patients participated in the study, with a mean age of 53.3 (SD = 13.0) years. Of these participants, 61.5 % were women. In CFA, five correlations between errors were included to fit the FACT-G to the sample (χ (2)/df = 8.611, CFI = .913, TLI = .902, RMSEA = .088). The model did not indicate invariant independent samples (Δχ(2): μ: p < .001, i: p < .958, Cov: p < .001, Res: p < .001). While there was adequate convergent validity for the physical well-being (AVE = .54) and social and family Well-being factors (AVE = .55), there was low convergent validity for the other factors. Reliability was adequate (CR = .76-.89 and α = .71-.82). Functional well-being, emotional well-being, and physical well-being were the factors that demonstrated a strong contribution to patients' health-related quality of life (β = -.99, .88, and .64, respectively). The FACT-G was found to be a valid and reliable assessment of health-related quality of life in a Brazilian sample of patients with cancer.
NASA Astrophysics Data System (ADS)
Moyer, P. A.; Boettcher, M. S.; McGuire, J. J.; Collins, J. A.
2015-12-01
On Gofar transform fault on the East Pacific Rise (EPR), Mw ~6.0 earthquakes occur every ~5 years and repeatedly rupture the same asperity (rupture patch), while the intervening fault segments (rupture barriers to the largest events) only produce small earthquakes. In 2008, an ocean bottom seismometer (OBS) deployment successfully captured the end of a seismic cycle, including an extensive foreshock sequence localized within a 10 km rupture barrier, the Mw 6.0 mainshock and its aftershocks that occurred in a ~10 km rupture patch, and an earthquake swarm located in a second rupture barrier. Here we investigate whether the inferred variations in frictional behavior along strike affect the rupture processes of 3.0 < M < 4.5 earthquakes by determining source parameters for 100 earthquakes recorded during the OBS deployment.Using waveforms with a 50 Hz sample rate from OBS accelerometers, we calculate stress drop using an omega-squared source model, where the weighted average corner frequency is derived from an empirical Green's function (EGF) method. We obtain seismic moment by fitting the omega-squared source model to the low frequency amplitude of individual spectra and account for attenuation using Q obtained from a velocity model through the foreshock zone. To ensure well-constrained corner frequencies, we require that the Brune [1970] model provides a statistically better fit to each spectral ratio than a linear model and that the variance is low between the data and model. To further ensure that the fit to the corner frequency is not influenced by resonance of the OBSs, we require a low variance close to the modeled corner frequency. Error bars on corner frequency were obtained through a grid search method where variance is within 10% of the best-fit value. Without imposing restrictive selection criteria, slight variations in corner frequencies from rupture patches and rupture barriers are not discernable. Using well-constrained source parameters, we find an average stress drop of 5.7 MPa in the aftershock zone compared to values of 2.4 and 2.9 MPa in the foreshock and swarm zones respectively. The higher stress drops in the rupture patch compared to the rupture barriers reflect systematic differences in along strike fault zone properties on Gofar transform fault.
Rice, Mabel L; Zubrick, Stephen R; Taylor, Catherine L; Gayán, Javier; Bontempo, Daniel E
2014-06-01
This study investigated the etiology of late language emergence (LLE) in 24-month-old twins, considering possible twinning, zygosity, gender, and heritability effects for vocabulary and grammar phenotypes. A population-based sample of 473 twin pairs participated. Multilevel modeling estimated means and variances of vocabulary and grammar phenotypes, controlling for familiality. Heritability was estimated with DeFries-Fulker regression and variance components models to determine effects of heritability, shared environment, and nonshared environment. Twins had lower average language scores than norms for single-born children, with lower average performance for monozygotic than dizygotic twins and for boys than girls, although gender and zygosity did not interact. Gender did not predict LLE. Significant heritability was detected for vocabulary (0.26) and grammar phenotypes (0.52 and 0.43 for boys and girls, respectively) in the full sample and in the sample selected for LLE (0.42 and 0.44). LLE and the appearance of Word Combinations were also significantly heritable (0.22-0.23). The findings revealed an increased likelihood of LLE in twin toddlers compared with single-born children that is modulated by zygosity and gender differences. Heritability estimates are consistent with previous research for vocabulary and add further suggestion of heritable differences in early grammar acquisition.
Model dependence and its effect on ensemble projections in CMIP5
NASA Astrophysics Data System (ADS)
Abramowitz, G.; Bishop, C.
2013-12-01
Conceptually, the notion of model dependence within climate model ensembles is relatively simple - modelling groups share a literature base, parametrisations, data sets and even model code - the potential for dependence in sampling different climate futures is clear. How though can this conceptual problem inform a practical solution that demonstrably improves the ensemble mean and ensemble variance as an estimate of system uncertainty? While some research has already focused on error correlation or error covariance as a candidate to improve ensemble mean estimates, a complete definition of independence must at least implicitly subscribe to an ensemble interpretation paradigm, such as the 'truth-plus-error', 'indistinguishable', or more recently 'replicate Earth' paradigm. Using a definition of model dependence based on error covariance within the replicate Earth paradigm, this presentation will show that accounting for dependence in surface air temperature gives cooler projections in CMIP5 - by as much as 20% globally in some RCPs - although results differ significantly for each RCP, especially regionally. The fact that the change afforded by accounting for dependence across different RCPs is different is not an inconsistent result. Different numbers of submissions to each RCP by different modelling groups mean that differences in projections from different RCPs are not entirely about RCP forcing conditions - they also reflect different sampling strategies.
Xie, Hongtu; Shi, Shaoying; Xiao, Hui; Xie, Chao; Wang, Feng; Fang, Qunle
2016-01-01
With the rapid development of the one-stationary bistatic forward-looking synthetic aperture radar (OS-BFSAR) technology, the huge amount of the remote sensing data presents challenges for real-time imaging processing. In this paper, an efficient time-domain algorithm (ETDA) considering the motion errors for the OS-BFSAR imaging processing, is presented. This method can not only precisely handle the large spatial variances, serious range-azimuth coupling and motion errors, but can also greatly improve the imaging efficiency compared with the direct time-domain algorithm (DTDA). Besides, it represents the subimages on polar grids in the ground plane instead of the slant-range plane, and derives the sampling requirements considering motion errors for the polar grids to offer a near-optimum tradeoff between the imaging precision and efficiency. First, OS-BFSAR imaging geometry is built, and the DTDA for the OS-BFSAR imaging is provided. Second, the polar grids of subimages are defined, and the subaperture imaging in the ETDA is derived. The sampling requirements for polar grids are derived from the point of view of the bandwidth. Finally, the implementation and computational load of the proposed ETDA are analyzed. Experimental results based on simulated and measured data validate that the proposed ETDA outperforms the DTDA in terms of the efficiency improvement. PMID:27845757
A refined method for multivariate meta-analysis and meta-regression.
Jackson, Daniel; Riley, Richard D
2014-02-20
Making inferences about the average treatment effect using the random effects model for meta-analysis is problematic in the common situation where there is a small number of studies. This is because estimates of the between-study variance are not precise enough to accurately apply the conventional methods for testing and deriving a confidence interval for the average effect. We have found that a refined method for univariate meta-analysis, which applies a scaling factor to the estimated effects' standard error, provides more accurate inference. We explain how to extend this method to the multivariate scenario and show that our proposal for refined multivariate meta-analysis and meta-regression can provide more accurate inferences than the more conventional approach. We explain how our proposed approach can be implemented using standard output from multivariate meta-analysis software packages and apply our methodology to two real examples. Copyright © 2013 John Wiley & Sons, Ltd.
Covariate Measurement Error Correction Methods in Mediation Analysis with Failure Time Data
Zhao, Shanshan
2014-01-01
Summary Mediation analysis is important for understanding the mechanisms whereby one variable causes changes in another. Measurement error could obscure the ability of the potential mediator to explain such changes. This paper focuses on developing correction methods for measurement error in the mediator with failure time outcomes. We consider a broad definition of measurement error, including technical error and error associated with temporal variation. The underlying model with the ‘true’ mediator is assumed to be of the Cox proportional hazards model form. The induced hazard ratio for the observed mediator no longer has a simple form independent of the baseline hazard function, due to the conditioning event. We propose a mean-variance regression calibration approach and a follow-up time regression calibration approach, to approximate the partial likelihood for the induced hazard function. Both methods demonstrate value in assessing mediation effects in simulation studies. These methods are generalized to multiple biomarkers and to both case-cohort and nested case-control sampling design. We apply these correction methods to the Women's Health Initiative hormone therapy trials to understand the mediation effect of several serum sex hormone measures on the relationship between postmenopausal hormone therapy and breast cancer risk. PMID:25139469
Covariate measurement error correction methods in mediation analysis with failure time data.
Zhao, Shanshan; Prentice, Ross L
2014-12-01
Mediation analysis is important for understanding the mechanisms whereby one variable causes changes in another. Measurement error could obscure the ability of the potential mediator to explain such changes. This article focuses on developing correction methods for measurement error in the mediator with failure time outcomes. We consider a broad definition of measurement error, including technical error, and error associated with temporal variation. The underlying model with the "true" mediator is assumed to be of the Cox proportional hazards model form. The induced hazard ratio for the observed mediator no longer has a simple form independent of the baseline hazard function, due to the conditioning event. We propose a mean-variance regression calibration approach and a follow-up time regression calibration approach, to approximate the partial likelihood for the induced hazard function. Both methods demonstrate value in assessing mediation effects in simulation studies. These methods are generalized to multiple biomarkers and to both case-cohort and nested case-control sampling designs. We apply these correction methods to the Women's Health Initiative hormone therapy trials to understand the mediation effect of several serum sex hormone measures on the relationship between postmenopausal hormone therapy and breast cancer risk. © 2014, The International Biometric Society.
NASA Technical Reports Server (NTRS)
Chang, Alfred T. C.; Chiu, Long S.; Wilheit, Thomas T.
1993-01-01
Global averages and random errors associated with the monthly oceanic rain rates derived from the Special Sensor Microwave/Imager (SSM/I) data using the technique developed by Wilheit et al. (1991) are computed. Accounting for the beam-filling bias, a global annual average rain rate of 1.26 m is computed. The error estimation scheme is based on the existence of independent (morning and afternoon) estimates of the monthly mean. Calculations show overall random errors of about 50-60 percent for each 5 deg x 5 deg box. The results are insensitive to different sampling strategy (odd and even days of the month). Comparison of the SSM/I estimates with raingage data collected at the Pacific atoll stations showed a low bias of about 8 percent, a correlation of 0.7, and an rms difference of 55 percent.
Xu, Stanley; Clarke, Christina L; Newcomer, Sophia R; Daley, Matthew F; Glanz, Jason M
2018-05-16
Vaccine safety studies are often electronic health record (EHR)-based observational studies. These studies often face significant methodological challenges, including confounding and misclassification of adverse event. Vaccine safety researchers use self-controlled case series (SCCS) study design to handle confounding effect and employ medical chart review to ascertain cases that are identified using EHR data. However, for common adverse events, limited resources often make it impossible to adjudicate all adverse events observed in electronic data. In this paper, we considered four approaches for analyzing SCCS data with confirmation rates estimated from an internal validation sample: (1) observed cases, (2) confirmed cases only, (3) known confirmation rate, and (4) multiple imputation (MI). We conducted a simulation study to evaluate these four approaches using type I error rates, percent bias, and empirical power. Our simulation results suggest that when misclassification of adverse events is present, approaches such as observed cases, confirmed case only, and known confirmation rate may inflate the type I error, yield biased point estimates, and affect statistical power. The multiple imputation approach considers the uncertainty of estimated confirmation rates from an internal validation sample, yields a proper type I error rate, largely unbiased point estimate, proper variance estimate, and statistical power. © 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
Mcalister, Courtney; Schmitter-Edgecombe, Maureen; Lamb, Richard
2016-03-01
The objective of this meta-analysis was to improve understanding of the heterogeneity in the relationship between cognition and functional status in individuals with mild cognitive impairment (MCI). Demographic, clinical, and methodological moderators were examined. Cognition explained an average of 23% of the variance in functional outcomes. Executive function measures explained the largest amount of variance (37%), whereas global cognitive status and processing speed measures explained the least (20%). Short- and long-delayed memory measures accounted for more variance (35% and 31%) than immediate memory measures (18%), and the relationship between cognition and functional outcomes was stronger when assessed with informant-report (28%) compared with self-report (21%). Demographics, sample characteristics, and type of everyday functioning measures (i.e., questionnaire, performance-based) explained relatively little variance compared with cognition. Executive functioning, particularly measured by Trails B, was a strong predictor of everyday functioning in individuals with MCI. A large proportion of variance remained unexplained by cognition. © The Author 2016. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Null-space and statistical significance of first-arrival traveltime inversion
NASA Astrophysics Data System (ADS)
Morozov, Igor B.
2004-03-01
The strong uncertainty inherent in the traveltime inversion of first arrivals from surface sources is usually removed by using a priori constraints or regularization. This leads to the null-space (data-independent model variability) being inadequately sampled, and consequently, model uncertainties may be underestimated in traditional (such as checkerboard) resolution tests. To measure the full null-space model uncertainties, we use unconstrained Monte Carlo inversion and examine the statistics of the resulting model ensembles. In an application to 1-D first-arrival traveltime inversion, the τ-p method is used to build a set of models that are equivalent to the IASP91 model within small, ~0.02 per cent, time deviations. The resulting velocity variances are much larger, ~2-3 per cent within the regions above the mantle discontinuities, and are interpreted as being due to the null-space. Depth-variant depth averaging is required for constraining the velocities within meaningful bounds, and the averaging scalelength could also be used as a measure of depth resolution. Velocity variances show structure-dependent, negative correlation with the depth-averaging scalelength. Neither the smoothest (Herglotz-Wiechert) nor the mean velocity-depth functions reproduce the discontinuities in the IASP91 model; however, the discontinuities can be identified by the increased null-space velocity (co-)variances. Although derived for a 1-D case, the above conclusions also relate to higher dimensions.
Derivation of an analytic expression for the error associated with the noise reduction rating
NASA Astrophysics Data System (ADS)
Murphy, William J.
2005-04-01
Hearing protection devices are assessed using the Real Ear Attenuation at Threshold (REAT) measurement procedure for the purpose of estimating the amount of noise reduction provided when worn by a subject. The rating number provided on the protector label is a function of the mean and standard deviation of the REAT results achieved by the test subjects. If a group of subjects have a large variance, then it follows that the certainty of the rating should be correspondingly lower. No estimate of the error of a protector's rating is given by existing standards or regulations. Propagation of errors was applied to the Noise Reduction Rating to develop an analytic expression for the hearing protector rating error term. Comparison of the analytic expression for the error to the standard deviation estimated from Monte Carlo simulation of subject attenuations yielded a linear relationship across several protector types and assumptions for the variance of the attenuations.
Spatial Variation of Soil Lead in an Urban Community Garden: Implications for Risk-Based Sampling.
Bugdalski, Lauren; Lemke, Lawrence D; McElmurry, Shawn P
2014-01-01
Soil lead pollution is a recalcitrant problem in urban areas resulting from a combination of historical residential, industrial, and transportation practices. The emergence of urban gardening movements in postindustrial cities necessitates accurate assessment of soil lead levels to ensure safe gardening. In this study, we examined small-scale spatial variability of soil lead within a 15 × 30 m urban garden plot established on two adjacent residential lots located in Detroit, Michigan, USA. Eighty samples collected using a variably spaced sampling grid were analyzed for total, fine fraction (less than 250 μm), and bioaccessible soil lead. Measured concentrations varied at sampling scales of 1-10 m and a hot spot exceeding 400 ppm total soil lead was identified in the northwest portion of the site. An interpolated map of total lead was treated as an exhaustive data set, and random sampling was simulated to generate Monte Carlo distributions and evaluate alternative sampling strategies intended to estimate the average soil lead concentration or detect hot spots. Increasing the number of individual samples decreases the probability of overlooking the hot spot (type II error). However, the practice of compositing and averaging samples decreased the probability of overestimating the mean concentration (type I error) at the expense of increasing the chance for type II error. The results reported here suggest a need to reconsider U.S. Environmental Protection Agency sampling objectives and consequent guidelines for reclaimed city lots where soil lead distributions are expected to be nonuniform. © 2013 Society for Risk Analysis.
A semiempirical error estimation technique for PWV derived from atmospheric radiosonde data
NASA Astrophysics Data System (ADS)
Castro-Almazán, Julio A.; Pérez-Jordán, Gabriel; Muñoz-Tuñón, Casiana
2016-09-01
A semiempirical method for estimating the error and optimum number of sampled levels in precipitable water vapour (PWV) determinations from atmospheric radiosoundings is proposed. Two terms have been considered: the uncertainties in the measurements and the sampling error. Also, the uncertainty has been separated in the variance and covariance components. The sampling and covariance components have been modelled from an empirical dataset of 205 high-vertical-resolution radiosounding profiles, equipped with Vaisala RS80 and RS92 sondes at four different locations: Güímar (GUI) in Tenerife, at sea level, and the astronomical observatory at Roque de los Muchachos (ORM, 2300 m a.s.l.) on La Palma (both on the Canary Islands, Spain), Lindenberg (LIN) in continental Germany, and Ny-Ålesund (NYA) in the Svalbard Islands, within the Arctic Circle. The balloons at the ORM were launched during intensive and unique site-testing runs carried out in 1990 and 1995, while the data for the other sites were obtained from radiosounding stations operating for a period of 1 year (2013-2014). The PWV values ranged between ˜ 0.9 and ˜ 41 mm. The method sub-samples the profile for error minimization. The result is the minimum error and the optimum number of levels. The results obtained in the four sites studied showed that the ORM is the driest of the four locations and the one with the fastest vertical decay of PWV. The exponential autocorrelation pressure lags ranged from 175 hPa (ORM) to 500 hPa (LIN). The results show a coherent behaviour with no biases as a function of the profile. The final error is roughly proportional to PWV whereas the optimum number of levels (N0) is the reverse. The value of N0 is less than 400 for 77 % of the profiles and the absolute errors are always < 0.6 mm. The median relative error is 2.0 ± 0.7 % and the 90th percentile P90 = 4.6 %. Therefore, whereas a radiosounding samples at least N0 uniform vertical levels, depending on the water vapour content and distribution of the atmosphere, the error in the PWV estimate is likely to stay below ≈ 3 %, even for dry conditions.
Environmental Influences on Well-Being: A Dyadic Latent Panel Analysis of Spousal Similarity
ERIC Educational Resources Information Center
Schimmack, Ulrich; Lucas, Richard E.
2010-01-01
This article uses dyadic latent panel analysis (DLPA) to examine environmental influences on well-being. DLPA requires longitudinal dyadic data. It decomposes the observed variance of both members of a dyad into a trait, state, and an error component. Furthermore, state variance is decomposed into initial and new state variance. Total observed…
Risk factors for near-miss events and safety incidents in pediatric radiation therapy.
Baig, Nimrah; Wang, Jiangxia; Elnahal, Shereef; McNutt, Todd; Wright, Jean; DeWeese, Theodore; Terezakis, Stephanie
2018-05-01
Factors contributing to safety- or quality-related incidents (e.g. variances) in children are unknown. We identified clinical and RT treatment variables associated with risk for variances in a pediatric cohort. Using our institution's incident learning system, 81 patients age ≤21 years old who experienced variances were compared to 191 pediatric patients without variances. Clinical and RT treatment variables were evaluated as potential predictors for variances using univariate and multivariate analyses. Variances were primarily documentation errors (n = 46, 57%) and were most commonly detected during treatment planning (n = 14, 21%). Treatment planning errors constituted the majority (n = 16 out of 29, 55%) of near-misses and safety incidents (NMSI), which excludes workflow incidents. Therapists reported the majority of variances (n = 50, 62%). Physician cross-coverage (OR = 2.1, 95% CI = 1.04-4.38) and 3D conformal RT (OR = 2.3, 95% CI = 1.11-4.69) increased variance risk. Conversely, age >14 years (OR = 0.5, 95% CI = 0.28-0.88) and diagnosis of abdominal tumor (OR = 0.2, 95% CI = 0.04-0.59) decreased variance risk. Variances in children occurred in early treatment phases, but were detected at later workflow stages. Quality measures should be implemented during early treatment phases with a focus on younger children and those cared for by cross-covering physicians. Copyright © 2018 Elsevier B.V. All rights reserved.
NASA Technical Reports Server (NTRS)
Osborne, William P.
1994-01-01
The use of 8 and 16 PSK TCM to support satellite communications in an effort to achieve more bandwidth efficiency in a power-limited channel has been proposed. This project addresses the problem of carrier phase jitter in an M-PSK receiver utilizing the high SNR approximation to the maximum aposteriori estimation of carrier phase. In particular, numerical solutions to the 8 and 16 PSK self-noise and phase detector gain in the carrier tracking loop are presented. The effect of changing SNR on the loop noise bandwidth is also discussed. These data are then used to compute variance of phase error as a function of SNR. Simulation and hardware data are used to verify these calculations. The results show that there is a threshold in the variance of phase error versus SNR curves that is a strong function of SNR and a weak function of loop bandwidth. The M-PSK variance thresholds occur at SNR's in the range of practical interest for the use of 8 and 16-PSK TCM. This suggests that phase error variance is an important consideration in the design of these systems.
Floyd A. Johnson
1961-01-01
This report assumes a knowledge of the principles of point sampling as described by Grosenbaugh, Bell and Alexander, and others. Whenever trees are counted at every point in a sample of points (large sample) and measured for volume at a portion (small sample) of these points, the sampling design could be called ratio double sampling. If the large...
NASA Astrophysics Data System (ADS)
Pan, X. G.; Wang, J. Q.; Zhou, H. Y.
2013-05-01
The variance component estimation (VCE) based on semi-parametric estimator with weighted matrix of data depth has been proposed, because the coupling system model error and gross error exist in the multi-source heterogeneous measurement data of space and ground combined TT&C (Telemetry, Tracking and Command) technology. The uncertain model error has been estimated with the semi-parametric estimator model, and the outlier has been restrained with the weighted matrix of data depth. On the basis of the restriction of the model error and outlier, the VCE can be improved and used to estimate weighted matrix for the observation data with uncertain model error or outlier. Simulation experiment has been carried out under the circumstance of space and ground combined TT&C. The results show that the new VCE based on the model error compensation can determine the rational weight of the multi-source heterogeneous data, and restrain the outlier data.
Determination of thorium by fluorescent x-ray spectrometry
Adler, I.; Axelrod, J.M.
1955-01-01
A fluorescent x-ray spectrographic method for the determination of thoria in rock samples uses thallium as an internal standard. Measurements are made with a two-channel spectrometer equipped with quartz (d = 1.817 A.) analyzing crystals. Particle-size effects are minimized by grinding the sample components with a mixture of silicon carbide and aluminum and then briquetting. Analyses of 17 samples showed that for the 16 samples containing over 0.7% thoria the average error, based on chemical results, is 4.7% and the maximum error, 9.5%. Because of limitations of instrumentation, 0.2% thoria is considered the lower limit of detection. An analysis can be made in about an hour.
2011-03-01
1.179 1 22 .289 POP-UP .000 1 22 .991 Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design ...POP-UP 2.104 1 22 .161 Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design : Intercept... design also limited the number of intended treatments. The experimental design originally was suppose to test all three adverse events that threaten
An Error-Reduction Algorithm to Improve Lidar Turbulence Estimates for Wind Energy
DOE Office of Scientific and Technical Information (OSTI.GOV)
Newman, Jennifer F.; Clifton, Andrew
2016-08-01
Currently, cup anemometers on meteorological (met) towers are used to measure wind speeds and turbulence intensity to make decisions about wind turbine class and site suitability. However, as modern turbine hub heights increase and wind energy expands to complex and remote sites, it becomes more difficult and costly to install met towers at potential sites. As a result, remote sensing devices (e.g., lidars) are now commonly used by wind farm managers and researchers to estimate the flow field at heights spanned by a turbine. While lidars can accurately estimate mean wind speeds and wind directions, there is still a largemore » amount of uncertainty surrounding the measurement of turbulence with lidars. This uncertainty in lidar turbulence measurements is one of the key roadblocks that must be overcome in order to replace met towers with lidars for wind energy applications. In this talk, a model for reducing errors in lidar turbulence estimates is presented. Techniques for reducing errors from instrument noise, volume averaging, and variance contamination are combined in the model to produce a corrected value of the turbulence intensity (TI), a commonly used parameter in wind energy. In the next step of the model, machine learning techniques are used to further decrease the error in lidar TI estimates.« less
NASA Technical Reports Server (NTRS)
Wolf, Michael
2012-01-01
A document describes an algorithm created to estimate the mass placed on a sample verification sensor (SVS) designed for lunar or planetary robotic sample return missions. A novel SVS measures the capacitance between a rigid bottom plate and an elastic top membrane in seven locations. As additional sample material (soil and/or small rocks) is placed on the top membrane, the deformation of the membrane increases the capacitance. The mass estimation algorithm addresses both the calibration of each SVS channel, and also addresses how to combine the capacitances read from each of the seven channels into a single mass estimate. The probabilistic approach combines the channels according to the variance observed during the training phase, and provides not only the mass estimate, but also a value for the certainty of the estimate. SVS capacitance data is collected for known masses under a wide variety of possible loading scenarios, though in all cases, the distribution of sample within the canister is expected to be approximately uniform. A capacitance-vs-mass curve is fitted to this data, and is subsequently used to determine the mass estimate for the single channel s capacitance reading during the measurement phase. This results in seven different mass estimates, one for each SVS channel. Moreover, the variance of the calibration data is used to place a Gaussian probability distribution function (pdf) around this mass estimate. To blend these seven estimates, the seven pdfs are combined into a single Gaussian distribution function, providing the final mean and variance of the estimate. This blending technique essentially takes the final estimate as an average of the estimates of the seven channels, weighted by the inverse of the channel s variance.
Estimating individual glomerular volume in the human kidney: clinical perspectives
Puelles, Victor G.; Zimanyi, Monika A.; Samuel, Terence; Hughson, Michael D.; Douglas-Denton, Rebecca N.; Bertram, John F.
2012-01-01
Background. Measurement of individual glomerular volumes (IGV) has allowed the identification of drivers of glomerular hypertrophy in subjects without overt renal pathology. This study aims to highlight the relevance of IGV measurements with possible clinical implications and determine how many profiles must be measured in order to achieve stable size distribution estimates. Methods. We re-analysed 2250 IGV estimates obtained using the disector/Cavalieri method in 41 African and 34 Caucasian Americans. Pooled IGV analysis of mean and variance was conducted. Monte-Carlo (Jackknife) simulations determined the effect of the number of sampled glomeruli on mean IGV. Lin’s concordance coefficient (RC), coefficient of variation (CV) and coefficient of error (CE) measured reliability. Results. IGV mean and variance increased with overweight and hypertensive status. Superficial glomeruli were significantly smaller than juxtamedullary glomeruli in all subjects (P < 0.01), by race (P < 0.05) and in obese individuals (P < 0.01). Subjects with multiple chronic kidney disease (CKD) comorbidities showed significant increases in IGV mean and variability. Overall, mean IGV was particularly reliable with nine or more sampled glomeruli (RC > 0.95, <5% difference in CV and CE). These observations were not affected by a reduced sample size and did not disrupt the inverse linear correlation between mean IGV and estimated total glomerular number. Conclusions. Multiple comorbidities for CKD are associated with increased IGV mean and variance within subjects, including overweight, obesity and hypertension. Zonal selection and the number of sampled glomeruli do not represent drawbacks for future longitudinal biopsy-based studies of glomerular size and distribution. PMID:21984554
Topographic analysis of individual activation patterns in medial frontal cortex in schizophrenia
Stern, Emily R.; Welsh, Robert C.; Fitzgerald, Kate D.; Taylor, Stephan F.
2009-01-01
Individual variability in the location of neural activations poses a unique problem for neuroimaging studies employing group averaging techniques to investigate the neural bases of cognitive and emotional functions. This may be especially challenging for studies examining patient groups, which often have limited sample sizes and increased intersubject variability. In particular, medial frontal cortex (MFC) dysfunction is thought to underlie performance monitoring dysfunction among patients with previous studies using group averaging to have yielded conflicting results. schizophrenia, yet compare schizophrenic patients to controls To examine individual activations in MFC associated with two aspects of performance monitoring, interference and error processing, functional magnetic resonance imaging (fMRI) data were acquired while 17 patients with schizophrenia and 21 healthy controls performed an event-related version of the multi-source interference task. Comparisons of averaged data revealed few differences between the groups. By contrast, topographic analysis of individual activations for errors showed that control subjects exhibited activations spanning across both posterior and anterior regions of MFC while patients primarily activated posterior MFC, possibly reflecting an impaired emotional response to errors in schizophrenia. This discrepancy between topographic and group-averaged results may be due to the significant dispersion among individual activations, particularly among healthy controls, highlighting the importance of considering intersubject variability when interpreting the medial frontal response to error commission. PMID:18819107
NASA Astrophysics Data System (ADS)
Kittiwisit, Piyanat; Bowman, Judd D.; Jacobs, Daniel C.; Beardsley, Adam P.; Thyagarajan, Nithyanandan
2018-03-01
We present a baseline sensitivity analysis of the Hydrogen Epoch of Reionization Array (HERA) and its build-out stages to one-point statistics (variance, skewness, and kurtosis) of redshifted 21 cm intensity fluctuation from the Epoch of Reionization (EoR) based on realistic mock observations. By developing a full-sky 21 cm light-cone model, taking into account the proper field of view and frequency bandwidth, utilizing a realistic measurement scheme, and assuming perfect foreground removal, we show that HERA will be able to recover statistics of the sky model with high sensitivity by averaging over measurements from multiple fields. All build-out stages will be able to detect variance, while skewness and kurtosis should be detectable for HERA128 and larger. We identify sample variance as the limiting constraint of the measurements at the end of reionization. The sensitivity can also be further improved by performing frequency windowing. In addition, we find that strong sample variance fluctuation in the kurtosis measured from an individual field of observation indicates the presence of outlying cold or hot regions in the underlying fluctuations, a feature that can potentially be used as an EoR bubble indicator.
Zucoloto, Miriane Lucindo; Maroco, João; Duarte Bonini Campos, Juliana Alvares
2015-01-01
To evaluate the psychometric properties of the Multidimensional Pain Inventory (MPI) in a Brazilian sample of patients with orofacial pain. A total of 1,925 adult patients, who sought dental care in the School of Dentistry of São Paulo State University's Araraquara campus, were invited to participate; 62.5% (n=1,203) agreed to participate. Of these, 436 presented with orofacial pain and were included. The mean age was 39.9 (SD=13.6) years and 74.5% were female. Confirmatory factor analysis was conducted using χ²/df, comparative fit index, goodness of fit index, and root mean square error of approximation as indices of goodness of fit. Convergent validity was estimated by the average variance extracted and composite reliability, and internal consistency by Cronbach's alpha standardized coefficient (α). The stability of the models was tested in independent samples (test and validation; dental pain and orofacial pain). The factorial invariance was estimated by multigroup analysis (Δχ²). Factorial, convergent validity, and internal consistency were adequate in all three parts of the MPI. To achieve this adequate fit for Part 1, item 15 needed to be deleted (λ=0.13). Discriminant validity was compromised between the factors "activities outside the home" and "social activities" of Part 3 of the MPI in the total sample, validation sample, and in patients with dental pain and with orofacial pain. A strong invariance between different subsamples from the three parts of the MPI was detected. The MPI produced valid, reliable, and stable data for pain assessment among Brazilian patients with orofacial pain.
Hard choices in assessing survival past dams — a comparison of single- and paired-release strategies
Zydlewski, Joseph D.; Stich, Daniel S.; Sigourney, Douglas B.
2017-01-01
Mark–recapture models are widely used to estimate survival of salmon smolts migrating past dams. Paired releases have been used to improve estimate accuracy by removing components of mortality not attributable to the dam. This method is accompanied by reduced precision because (i) sample size is reduced relative to a single, large release; and (ii) variance calculations inflate error. We modeled an idealized system with a single dam to assess trade-offs between accuracy and precision and compared methods using root mean squared error (RMSE). Simulations were run under predefined conditions (dam mortality, background mortality, detection probability, and sample size) to determine scenarios when the paired release was preferable to a single release. We demonstrate that a paired-release design provides a theoretical advantage over a single-release design only at large sample sizes and high probabilities of detection. At release numbers typical of many survival studies, paired release can result in overestimation of dam survival. Failures to meet model assumptions of a paired release may result in further overestimation of dam-related survival. Under most conditions, a single-release strategy was preferable.
(Sample) Size Matters: Best Practices for Defining Error in Planktic Foraminiferal Proxy Records
NASA Astrophysics Data System (ADS)
Lowery, C.; Fraass, A. J.
2016-02-01
Paleoceanographic research is a vital tool to extend modern observational datasets and to study the impact of climate events for which there is no modern analog. Foraminifera are one of the most widely used tools for this type of work, both as paleoecological indicators and as carriers for geochemical proxies. However, the use of microfossils as proxies for paleoceanographic conditions brings about a unique set of problems. This is primarily due to the fact that groups of individual foraminifera, which usually live about a month, are used to infer average conditions for time periods ranging from hundreds to tens of thousands of years. Because of this, adequate sample size is very important for generating statistically robust datasets, particularly for stable isotopes. In the early days of stable isotope geochemistry, instrumental limitations required hundreds of individual foraminiferal tests to return a value. This had the fortunate side-effect of smoothing any seasonal to decadal changes within the planktic foram population. With the advent of more sensitive mass spectrometers, smaller sample sizes have now become standard. While this has many advantages, the use of smaller numbers of individuals to generate a data point has lessened the amount of time averaging in the isotopic analysis and decreased precision in paleoceanographic datasets. With fewer individuals per sample, the differences between individual specimens will result in larger variation, and therefore error, and less precise values for each sample. Unfortunately, most (the authors included) do not make a habit of reporting the error associated with their sample size. We have created an open-source model in R to quantify the effect of sample sizes under various realistic and highly modifiable parameters (calcification depth, diagenesis in a subset of the population, improper identification, vital effects, mass, etc.). For example, a sample in which only 1 in 10 specimens is diagenetically altered can be off by >0.3‰ δ18O VPDB, or 1°C. Here, we demonstrate the use of this tool to quantify error in micropaleontological datasets, and suggest best practices for minimizing error when generating stable isotope data with foraminifera.
Lack of consensus among competency ratings of the same occupation: noise or substance?
Lievens, Filip; Sanchez, Juan I; Bartram, Dave; Brown, Anna
2010-05-01
Although rating differences among incumbents of the same occupation have traditionally been viewed as error variance in the work analysis domain, such differences might often capture substantive discrepancies in how incumbents approach their work. This study draws from job crafting, creativity, and role theories to uncover situational factors (i.e., occupational activities, context, and complexity) related to differences among competency ratings of the same occupation. The sample consisted of 192 incumbents from 64 occupations. Results showed that 25% of the variance associated with differences in competency ratings of the same occupation was related to the complexity, the context, and primarily the nature of the occupation's work activities. Consensus was highest for occupations involving equipment-related activities and direct contact with the public. PsycINFO Database Record (c) 2010 APA, all rights reserved.
Jin, Mengtong; Sun, Wenshuo; Li, Qin; Sun, Xiaohong; Pan, Yingjie; Zhao, Yong
2014-04-04
We evaluated the difference of three standard curves in quantifying viable Vibrio parahaemolyticus in samples by real-time reverse-transcriptase PCR (Real-time RT-PCR). The standard curve A was established by 10-fold diluted cDNA. The cDNA was reverse transcripted after RNA synthesized in vitro. The standard curve B and C were established by 10-fold diluted cDNA. The cDNA was synthesized after RNA isolated from Vibrio parahaemolyticus in pure cultures (10(8) CFU/mL) and shrimp samples (10(6) CFU/g) (Standard curve A and C were proposed for the first time). Three standard curves were performed to quantitatively detect V. parahaemolyticus in six samples, respectively (Two pure cultured V. parahaemolyticus samples, two artificially contaminated cooked Litopenaeus vannamei samples and two artificially contaminated Litopenaeus vannamei samples). Then we evaluated the quantitative results of standard curve and the plate counting results and then analysed the differences. The three standard curves all show a strong linear relationship between the fractional cycle number and V. parahaemolyticus concentration (R2 > 0.99); The quantitative results of Real-time PCR were significantly (p < 0.05) lower than the results of plate counting. The relative errors compared with the results of plate counting ranked standard curve A (30.0%) > standard curve C (18.8%) > standard curve B (6.9%); The average differences between standard curve A and standard curve B and C were - 2.25 Lg CFU/mL and - 0.75 Lg CFU/mL, respectively, and the mean relative errors were 48.2% and 15.9%, respectively; The average difference between standard curve B and C was among (1.47 -1.53) Lg CFU/mL and the average relative errors were among 19.0% - 23.8%. Standard curve B could be applied to Real-time RT-PCR when quantify the number of viable microorganisms in samples.
Junttila, Virpi; Kauranne, Tuomo; Finley, Andrew O.; Bradford, John B.
2015-01-01
Modern operational forest inventory often uses remotely sensed data that cover the whole inventory area to produce spatially explicit estimates of forest properties through statistical models. The data obtained by airborne light detection and ranging (LiDAR) correlate well with many forest inventory variables, such as the tree height, the timber volume, and the biomass. To construct an accurate model over thousands of hectares, LiDAR data must be supplemented with several hundred field sample measurements of forest inventory variables. This can be costly and time consuming. Different LiDAR-data-based and spatial-data-based sampling designs can reduce the number of field sample plots needed. However, problems arising from the features of the LiDAR data, such as a large number of predictors compared with the sample size (overfitting) or a strong correlation among predictors (multicollinearity), may decrease the accuracy and precision of the estimates and predictions. To overcome these problems, a Bayesian linear model with the singular value decomposition of predictors, combined with regularization, is proposed. The model performance in predicting different forest inventory variables is verified in ten inventory areas from two continents, where the number of field sample plots is reduced using different sampling designs. The results show that, with an appropriate field plot selection strategy and the proposed linear model, the total relative error of the predicted forest inventory variables is only 5%–15% larger using 50 field sample plots than the error of a linear model estimated with several hundred field sample plots when we sum up the error due to both the model noise variance and the model’s lack of fit.
Estimating Variances of Horizontal Wind Fluctuations in Stable Conditions
NASA Astrophysics Data System (ADS)
Luhar, Ashok K.
2010-05-01
Information concerning the average wind speed and the variances of lateral and longitudinal wind velocity fluctuations is required by dispersion models to characterise turbulence in the atmospheric boundary layer. When the winds are weak, the scalar average wind speed and the vector average wind speed need to be clearly distinguished and both lateral and longitudinal wind velocity fluctuations assume equal importance in dispersion calculations. We examine commonly-used methods of estimating these variances from wind-speed and wind-direction statistics measured separately, for example, by a cup anemometer and a wind vane, and evaluate the implied relationship between the scalar and vector wind speeds, using measurements taken under low-wind stable conditions. We highlight several inconsistencies inherent in the existing formulations and show that the widely-used assumption that the lateral velocity variance is equal to the longitudinal velocity variance is not necessarily true. We derive improved relations for the two variances, and although data under stable stratification are considered for comparison, our analysis is applicable more generally.
NASA Astrophysics Data System (ADS)
Rock, N. M. S.; Duffy, T. R.
REGRES allows a range of regression equations to be calculated for paired sets of data values in which both variables are subject to error (i.e. neither is the "independent" variable). Nonparametric regressions, based on medians of all possible pairwise slopes and intercepts, are treated in detail. Estimated slopes and intercepts are output, along with confidence limits, Spearman and Kendall rank correlation coefficients. Outliers can be rejected with user-determined stringency. Parametric regressions can be calculated for any value of λ (the ratio of the variances of the random errors for y and x)—including: (1) major axis ( λ = 1); (2) reduced major axis ( λ = variance of y/variance of x); (3) Y on Xλ = infinity; or (4) X on Y ( λ = 0) solutions. Pearson linear correlation coefficients also are output. REGRES provides an alternative to conventional isochron assessment techniques where bivariate normal errors cannot be assumed, or weighting methods are inappropriate.
Linking models and data on vegetation structure
NASA Astrophysics Data System (ADS)
Hurtt, G. C.; Fisk, J.; Thomas, R. Q.; Dubayah, R.; Moorcroft, P. R.; Shugart, H. H.
2010-06-01
For more than a century, scientists have recognized the importance of vegetation structure in understanding forest dynamics. Now future satellite missions such as Deformation, Ecosystem Structure, and Dynamics of Ice (DESDynI) hold the potential to provide unprecedented global data on vegetation structure needed to reduce uncertainties in terrestrial carbon dynamics. Here, we briefly review the uses of data on vegetation structure in ecosystem models, develop and analyze theoretical models to quantify model-data requirements, and describe recent progress using a mechanistic modeling approach utilizing a formal scaling method and data on vegetation structure to improve model predictions. Generally, both limited sampling and coarse resolution averaging lead to model initialization error, which in turn is propagated in subsequent model prediction uncertainty and error. In cases with representative sampling, sufficient resolution, and linear dynamics, errors in initialization tend to compensate at larger spatial scales. However, with inadequate sampling, overly coarse resolution data or models, and nonlinear dynamics, errors in initialization lead to prediction error. A robust model-data framework will require both models and data on vegetation structure sufficient to resolve important environmental gradients and tree-level heterogeneity in forest structure globally.
Evaluation of JGM 2 geopotential errors from geosat, TOPEX/poseidon and ERS-1 crossover altimetry
NASA Astrophysics Data System (ADS)
Wagner, C. A.; Klokocník, J.; Tai, C. K.
1995-08-01
World-ocean distribution of the crossover altimetry data from Geosat, TOPEX/Poseidon (T/P) and the ERS 1 missions have provided strong independent evidence that NASA's/CSR's JGM 2 geopotential model (70 x 70 in spherical harmonics) yields accurate radial ephemerides for these satellites. In testing the sea height crossover differences found from altimetry and JGM 2 orbits for these satellites, we have used the sea height differences themselves (of ascending minus descending passes averaged at each location over many exact repeat cycles) and the Lumped Latitude Coefficients (LLC) derived from them. For Geosat we find the geopotential-induced LLC errors (exclusive of non-gravitational and initial state discrepancies) mostly below 6 cm, for TOPEX the corresponding errors are usually below 2 cm, and for ERS 1 (35-day cycle) they are generally belo2 5 cm. In addition, we have found that these observations agree well overall with predictions of accuracy derived from the JGM 2 variance-covariance matrix; the corresponding projected LLC errors for Geosat, T/P, and ERS 1 are usually between 1 and 4 cm, 1 - 2 cm, and 1 - 4 cm, respectively (they depend on the filtering of long-periodic perturbations and on the order of the LLC). This agreement is especially impressive for ERS 1 since no data of any kind from this mission was used in forming JGM 2. The observed crossover differences for Geosat, T/P and ERS 1 are 8, 3, and 11 cm (rms), respectively. These observations also agree well with prediction of accuracy derived from the JGM 2 variance-covariance matrix; the corresponding projected crossover errors for Geosat and T/P are 8 cm and 2.3 cm, respectively. The precision of our mean difference observations is about 3 cm for Geosat (approx. 24,000 observations), 1.5 cm for T/P (approx. 6,000 observations) and 5 cm for ERS 1 (approx. 44,000 observations). Thus, these ``global'' independent data should provide a valuable new source for improving geopotential models. Our results show the need for further correction of the low order JGM 2 geopotential as well as certain resonant orders for all 3 satellites.
ERIC Educational Resources Information Center
Abry, Tashia; Cash, Anne H.; Bradshaw, Catherine P.
2014-01-01
Generalizability theory (GT) offers a useful framework for estimating the reliability of a measure while accounting for multiple sources of error variance. The purpose of this study was to use GT to examine multiple sources of variance in and the reliability of school-level teacher and high school student behaviors as observed using the tool,…
ERIC Educational Resources Information Center
Lix, Lisa M.; And Others
1996-01-01
Meta-analytic techniques were used to summarize the statistical robustness literature on Type I error properties of alternatives to the one-way analysis of variance "F" test. The James (1951) and Welch (1951) tests performed best under violations of the variance homogeneity assumption, although their use is not always appropriate. (SLD)
Network Sampling with Memory: A proposal for more efficient sampling from social networks.
Mouw, Ted; Verdery, Ashton M
2012-08-01
Techniques for sampling from networks have grown into an important area of research across several fields. For sociologists, the possibility of sampling from a network is appealing for two reasons: (1) A network sample can yield substantively interesting data about network structures and social interactions, and (2) it is useful in situations where study populations are difficult or impossible to survey with traditional sampling approaches because of the lack of a sampling frame. Despite its appeal, methodological concerns about the precision and accuracy of network-based sampling methods remain. In particular, recent research has shown that sampling from a network using a random walk based approach such as Respondent Driven Sampling (RDS) can result in high design effects (DE)-the ratio of the sampling variance to the sampling variance of simple random sampling (SRS). A high design effect means that more cases must be collected to achieve the same level of precision as SRS. In this paper we propose an alternative strategy, Network Sampling with Memory (NSM), which collects network data from respondents in order to reduce design effects and, correspondingly, the number of interviews needed to achieve a given level of statistical power. NSM combines a "List" mode, where all individuals on the revealed network list are sampled with the same cumulative probability, with a "Search" mode, which gives priority to bridge nodes connecting the current sample to unexplored parts of the network. We test the relative efficiency of NSM compared to RDS and SRS on 162 school and university networks from Add Health and Facebook that range in size from 110 to 16,278 nodes. The results show that the average design effect for NSM on these 162 networks is 1.16, which is very close to the efficiency of a simple random sample (DE=1), and 98.5% lower than the average DE we observed for RDS.
Network Sampling with Memory: A proposal for more efficient sampling from social networks
Mouw, Ted; Verdery, Ashton M.
2013-01-01
Techniques for sampling from networks have grown into an important area of research across several fields. For sociologists, the possibility of sampling from a network is appealing for two reasons: (1) A network sample can yield substantively interesting data about network structures and social interactions, and (2) it is useful in situations where study populations are difficult or impossible to survey with traditional sampling approaches because of the lack of a sampling frame. Despite its appeal, methodological concerns about the precision and accuracy of network-based sampling methods remain. In particular, recent research has shown that sampling from a network using a random walk based approach such as Respondent Driven Sampling (RDS) can result in high design effects (DE)—the ratio of the sampling variance to the sampling variance of simple random sampling (SRS). A high design effect means that more cases must be collected to achieve the same level of precision as SRS. In this paper we propose an alternative strategy, Network Sampling with Memory (NSM), which collects network data from respondents in order to reduce design effects and, correspondingly, the number of interviews needed to achieve a given level of statistical power. NSM combines a “List” mode, where all individuals on the revealed network list are sampled with the same cumulative probability, with a “Search” mode, which gives priority to bridge nodes connecting the current sample to unexplored parts of the network. We test the relative efficiency of NSM compared to RDS and SRS on 162 school and university networks from Add Health and Facebook that range in size from 110 to 16,278 nodes. The results show that the average design effect for NSM on these 162 networks is 1.16, which is very close to the efficiency of a simple random sample (DE=1), and 98.5% lower than the average DE we observed for RDS. PMID:24159246
The role of global cloud climatologies in validating numerical models
NASA Technical Reports Server (NTRS)
HARSHVARDHAN
1993-01-01
The purpose of this work is to estimate sampling errors of area-time averaged rain rate due to temporal samplings by satellites. In particular, the sampling errors of the proposed low inclination orbit satellite of the Tropical Rainfall Measuring Mission (TRMM) (35 deg inclination and 350 km altitude), one of the sun synchronous polar orbiting satellites of NOAA series (98.89 deg inclination and 833 km altitude), and two simultaneous sun synchronous polar orbiting satellites--assumed to carry a perfect passive microwave sensor for direct rainfall measurements--will be estimated. This estimate is done by performing a study of the satellite orbits and the autocovariance function of the area-averaged rain rate time series. A model based on an exponential fit of the autocovariance function is used for actual calculations. Varying visiting intervals and total coverage of averaging area on each visit by the satellites are taken into account in the model. The data are generated by a General Circulation Model (GCM). The model has a diurnal cycle and parameterized convective processes. A special run of the GCM was made at NASA/GSFC in which the rainfall and precipitable water fields were retained globally for every hour of the run for the whole year.
High-frequency signal and noise estimates of CSR GRACE RL04
NASA Astrophysics Data System (ADS)
Bonin, Jennifer A.; Bettadpur, Srinivas; Tapley, Byron D.
2012-12-01
A sliding window technique is used to create daily-sampled Gravity Recovery and Climate Experiment (GRACE) solutions with the same background processing as the official CSR RL04 monthly series. By estimating over shorter time spans, more frequent solutions are made using uncorrelated data, allowing for higher frequency resolution in addition to daily sampling. Using these data sets, high-frequency GRACE errors are computed using two different techniques: assuming the GRACE high-frequency signal in a quiet area of the ocean is the true error, and computing the variance of differences between multiple high-frequency GRACE series from different centers. While the signal-to-noise ratios prove to be sufficiently high for confidence at annual and lower frequencies, at frequencies above 3 cycles/year the signal-to-noise ratios in the large hydrological basins looked at here are near 1.0. Comparisons with the GLDAS hydrological model and high frequency GRACE series developed at other centers confirm CSR GRACE RL04's poor ability to accurately and reliably measure hydrological signal above 3-9 cycles/year, due to the low power of the large-scale hydrological signal typical at those frequencies compared to the GRACE errors.
Statistical Analyses of Scatterplots to Identify Important Factors in Large-Scale Simulations
DOE Office of Scientific and Technical Information (OSTI.GOV)
Kleijnen, J.P.C.; Helton, J.C.
1999-04-01
The robustness of procedures for identifying patterns in scatterplots generated in Monte Carlo sensitivity analyses is investigated. These procedures are based on attempts to detect increasingly complex patterns in the scatterplots under consideration and involve the identification of (1) linear relationships with correlation coefficients, (2) monotonic relationships with rank correlation coefficients, (3) trends in central tendency as defined by means, medians and the Kruskal-Wallis statistic, (4) trends in variability as defined by variances and interquartile ranges, and (5) deviations from randomness as defined by the chi-square statistic. The following two topics related to the robustness of these procedures are consideredmore » for a sequence of example analyses with a large model for two-phase fluid flow: the presence of Type I and Type II errors, and the stability of results obtained with independent Latin hypercube samples. Observations from analysis include: (1) Type I errors are unavoidable, (2) Type II errors can occur when inappropriate analysis procedures are used, (3) physical explanations should always be sought for why statistical procedures identify variables as being important, and (4) the identification of important variables tends to be stable for independent Latin hypercube samples.« less
Estimating the theoretical semivariogram from finite numbers of measurements
Zheng, Li; Silliman, Stephen E.
2000-01-01
We investigate from a theoretical basis the impacts of the number, location, and correlation among measurement points on the quality of an estimate of the semivariogram. The unbiased nature of the semivariogram estimator ŷ(r) is first established for a general random process Z(x). The variance of ŷZ(r) is then derived as a function of the sampling parameters (the number of measurements and their locations). In applying this function to the case of estimating the semivariograms of the transmissivity and the hydraulic head field, it is shown that the estimation error depends on the number of the data pairs, the correlation among the data pairs (which, in turn, are determined by the form of the underlying semivariogram γ(r)), the relative locations of the data pairs, and the separation distance at which the semivariogram is to be estimated. Thus design of an optimal sampling program for semivariogram estimation should include consideration of each of these factors. Further, the function derived for the variance of ŷZ(r) is useful in determining the reliability of a semivariogram developed from a previously established sampling design.
Vaskinn, Anja; Andersson, Stein; Østefjells, Tiril; Andreassen, Ole A; Sundet, Kjetil
2018-06-05
Theory of mind (ToM) can be divided into cognitive and affective ToM, and a distinction can be made between overmentalizing and undermentalizing errors. Research has shown that ToM in schizophrenia is associated with non-social and social cognition, and with clinical symptoms. In this study, we investigate cognitive and clinical predictors of different ToM processes. Ninety-one individuals with schizophrenia participated. ToM was measured with the Movie for the Assessment of Social Cognition (MASC) yielding six scores (total ToM, cognitive ToM, affective ToM, overmentalizing errors, undermentalizing errors and no mentalizing errors). Neurocognition was indexed by a composite score based on the non-social cognitive tests in the MATRICS Consensus Cognitive Battery (MCCB). Emotion perception was measured with Emotion in Biological Motion (EmoBio), a point-light walker task. Clinical symptoms were assessed with the Positive and Negative Syndrome Scale (PANSS). Seventy-one healthy control (HC) participants completed the MASC. Individuals with schizophrenia showed large impairments compared to HC for all MASC scores, except overmentalizing errors. Hierarchical regression analyses with the six different MASC scores as dependent variables revealed that MCCB was a significant predictor of all MASC scores, explaining 8-18% of the variance. EmoBio increased the explained variance significantly, to 17-28%, except for overmentalizing errors. PANSS excited symptoms increased explained variance for total ToM, affective ToM and no mentalizing errors. Both social and non-social cognition were significant predictors of ToM. Overmentalizing was only predicted by non-social cognition. Excited symptoms contributed to overall and affective ToM, and to no mentalizing errors. Copyright © 2018 Elsevier Inc. All rights reserved.
Evaluating causes of error in landmark-based data collection using scanners
Shearer, Brian M.; Cooke, Siobhán B.; Halenar, Lauren B.; Reber, Samantha L.; Plummer, Jeannette E.; Delson, Eric
2017-01-01
In this study, we assess the precision, accuracy, and repeatability of craniodental landmarks (Types I, II, and III, plus curves of semilandmarks) on a single macaque cranium digitally reconstructed with three different surface scanners and a microCT scanner. Nine researchers with varying degrees of osteological and geometric morphometric knowledge landmarked ten iterations of each scan (40 total) to test the effects of scan quality, researcher experience, and landmark type on levels of intra- and interobserver error. Two researchers additionally landmarked ten specimens from seven different macaque species using the same landmark protocol to test the effects of the previously listed variables relative to species-level morphological differences (i.e., observer variance versus real biological variance). Error rates within and among researchers by scan type were calculated to determine whether or not data collected by different individuals or on different digitally rendered crania are consistent enough to be used in a single dataset. Results indicate that scan type does not impact rate of intra- or interobserver error. Interobserver error is far greater than intraobserver error among all individuals, and is similar in variance to that found among different macaque species. Additionally, experience with osteology and morphometrics both positively contribute to precision in multiple landmarking sessions, even where less experienced researchers have been trained in point acquisition. Individual training increases precision (although not necessarily accuracy), and is highly recommended in any situation where multiple researchers will be collecting data for a single project. PMID:29099867
Range camera on conveyor belts: estimating size distribution and systematic errors due to occlusion
NASA Astrophysics Data System (ADS)
Blomquist, Mats; Wernersson, Ake V.
1999-11-01
When range cameras are used for analyzing irregular material on a conveyor belt there will be complications like missing segments caused by occlusion. Also, a number of range discontinuities will be present. In a frame work towards stochastic geometry, conditions are found for the cases when range discontinuities take place. The test objects in this paper are pellets for the steel industry. An illuminating laser plane will give range discontinuities at the edges of each individual object. These discontinuities are used to detect and measure the chord created by the intersection of the laser plane and the object. From the measured chords we derive the average diameter and its variance. An improved method is to use a pair of parallel illuminating light planes to extract two chords. The estimation error for this method is not larger than the natural shape fluctuations (the difference in diameter) for the pellets. The laser- camera optronics is sensitive enough both for material on a conveyor belt and free falling material leaving the conveyor.
Schulze, P.A.; Capel, P.D.; Squillace, P.J.; Helsel, D.R.
1993-01-01
The usefulness and sensitivity, of a portable immunoassay test for the semiquantitative field screening of water samples was evaluated by means of laboratory and field studies. Laboratory results indicated that the tests were useful for the determination of atrazine concentrations of 0.1 to 1.5 μg/L. At a concentration of 1 μg/L, the relative standard deviation in the difference between the regression line and the actual result was about 40 percent. The immunoassay was less sensitive and produced similar errors for other triazine herbicides. After standardization, the test results were relatively insensitive to ionic content and variations in pH (range, 4 to 10), mildly sensitive to temperature changes, and quite sensitive to the timing of the final incubation step, variances in timing can be a significant source of error. Almost all of the immunoassays predicted a higher atrazine concentration in water samples when compared to results of gas chromatography. If these tests are used as a semiquantitative screening tool, this tendency for overprediction does not diminish the tests' usefulness. Generally, the tests seem to be a valuable method for screening water samples for triazine herbicides.
Badrick, Tony; Graham, Peter
2018-03-28
Internal Quality Control and External Quality Assurance are separate but related processes that have developed independently in laboratory medicine over many years. They have different sample frequencies, statistical interpretations and immediacy. Both processes have evolved absorbing new understandings of the concept of laboratory error, sample material matrix and assay capability. However, we do not believe at the coalface that either process has led to much improvement in patient outcomes recently. It is the increasing reliability and automation of analytical platforms along with improved stability of reagents that has reduced systematic and random error, which in turn has minimised the risk of running less frequent IQC. We suggest that it is time to rethink the role of both these processes and unite them into a single approach using an Average of Normals model supported by more frequent External Quality Assurance samples. This new paradigm may lead to less confusion for laboratory staff and quicker responses to and identification of out of control situations.
Lee, It Ee; Ghassemlooy, Zabih; Ng, Wai Pang; Khalighi, Mohammad-Ali; Liaw, Shien-Kuei
2016-01-01
Joint effects of aperture averaging and beam width on the performance of free-space optical communication links, under the impairments of atmospheric loss, turbulence, and pointing errors (PEs), are investigated from an information theory perspective. The propagation of a spatially partially coherent Gaussian-beam wave through a random turbulent medium is characterized, taking into account the diverging and focusing properties of the optical beam as well as the scintillation and beam wander effects. Results show that a noticeable improvement in the average channel capacity can be achieved with an enlarged receiver aperture in the moderate-to-strong turbulence regime, even without knowledge of the channel state information. In particular, it is observed that the optimum beam width can be reduced to improve the channel capacity, albeit the presence of scintillation and PEs, given that either one or both of these adverse effects are least dominant. We show that, under strong turbulence conditions, the beam width increases linearly with the Rytov variance for a relatively smaller PE loss but changes exponentially with steeper increments for higher PE losses. Our findings conclude that the optimal beam width is dependent on the combined effects of turbulence and PEs, and this parameter should be adjusted according to the varying atmospheric channel conditions. Therefore, we demonstrate that the maximum channel capacity is best achieved through the introduction of a larger receiver aperture and a beam-width optimization technique.
Why GPS makes distances bigger than they are
Ranacher, Peter; Brunauer, Richard; Trutschnig, Wolfgang; Van der Spek, Stefan; Reich, Siegfried
2016-01-01
ABSTRACT Global navigation satellite systems such as the Global Positioning System (GPS) is one of the most important sensors for movement analysis. GPS is widely used to record the trajectories of vehicles, animals and human beings. However, all GPS movement data are affected by both measurement and interpolation errors. In this article we show that measurement error causes a systematic bias in distances recorded with a GPS; the distance between two points recorded with a GPS is – on average – bigger than the true distance between these points. This systematic ‘overestimation of distance’ becomes relevant if the influence of interpolation error can be neglected, which in practice is the case for movement sampled at high frequencies. We provide a mathematical explanation of this phenomenon and illustrate that it functionally depends on the autocorrelation of GPS measurement error (C). We argue that C can be interpreted as a quality measure for movement data recorded with a GPS. If there is a strong autocorrelation between any two consecutive position estimates, they have very similar error. This error cancels out when average speed, distance or direction is calculated along the trajectory. Based on our theoretical findings we introduce a novel approach to determine C in real-world GPS movement data sampled at high frequencies. We apply our approach to pedestrian trajectories and car trajectories. We found that the measurement error in the data was strongly spatially and temporally autocorrelated and give a quality estimate of the data. Most importantly, our findings are not limited to GPS alone. The systematic bias and its implications are bound to occur in any movement data collected with absolute positioning if interpolation error can be neglected. PMID:27019610
On Time/Space Aggregation of Fine-Scale Error Estimates (Invited)
NASA Astrophysics Data System (ADS)
Huffman, G. J.
2013-12-01
Estimating errors inherent in fine time/space-scale satellite precipitation data sets is still an on-going problem and a key area of active research. Complicating features of these data sets include the intrinsic intermittency of the precipitation in space and time and the resulting highly skewed distribution of precipitation rates. Additional issues arise from the subsampling errors that satellites introduce, the errors due to retrieval algorithms, and the correlated error that retrieval and merger algorithms sometimes introduce. Several interesting approaches have been developed recently that appear to make progress on these long-standing issues. At the same time, the monthly averages over 2.5°x2.5° grid boxes in the Global Precipitation Climatology Project (GPCP) Satellite-Gauge (SG) precipitation data set follow a very simple sampling-based error model (Huffman 1997) with coefficients that are set using coincident surface and GPCP SG data. This presentation outlines the unsolved problem of how to aggregate the fine-scale errors (discussed above) to an arbitrary time/space averaging volume for practical use in applications, reducing in the limit to simple Gaussian expressions at the monthly 2.5°x2.5° scale. Scatter diagrams with different time/space averaging show that the relationship between the satellite and validation data improves due to the reduction in random error. One of the key, and highly non-linear, issues is that fine-scale estimates tend to have large numbers of cases with points near the axes on the scatter diagram (one of the values is exactly or nearly zero, while the other value is higher). Averaging 'pulls' the points away from the axes and towards the 1:1 line, which usually happens for higher precipitation rates before lower rates. Given this qualitative observation of how aggregation affects error, we observe that existing aggregation rules, such as the Steiner et al. (2003) power law, only depend on the aggregated precipitation rate. Is this sufficient, or is it necessary to aggregate the precipitation error estimates across the time/space data cube used for averaging? At least for small time/space data cubes it would seem that the detailed variables that affect each precipitation error estimate in the aggregation, such as sensor type, land/ocean surface type, convective/stratiform type, and so on, drive variations that must be accounted for explicitly.
Development and Evaluation of Algorithms for Breath Alcohol Screening.
Ljungblad, Jonas; Hök, Bertil; Ekström, Mikael
2016-04-01
Breath alcohol screening is important for traffic safety, access control and other areas of health promotion. A family of sensor devices useful for these purposes is being developed and evaluated. This paper is focusing on algorithms for the determination of breath alcohol concentration in diluted breath samples using carbon dioxide to compensate for the dilution. The examined algorithms make use of signal averaging, weighting and personalization to reduce estimation errors. Evaluation has been performed by using data from a previously conducted human study. It is concluded that these features in combination will significantly reduce the random error compared to the signal averaging algorithm taken alone.
Flanders, W Dana; Kirkland, Kimberly H; Shelton, Brian G
2014-10-01
Outbreaks of Legionnaires' disease require environmental testing of water samples from potentially implicated building water systems to identify the source of exposure. A previous study reports a large impact on Legionella sample results due to shipping and delays in sample processing. Specifically, this same study, without accounting for measurement error, reports more than half of shipped samples tested had Legionella levels that arbitrarily changed up or down by one or more logs, and the authors attribute this result to shipping time. Accordingly, we conducted a study to determine the effects of sample holding/shipping time on Legionella sample results while taking into account measurement error, which has previously not been addressed. We analyzed 159 samples, each split into 16 aliquots, of which one-half (8) were processed promptly after collection. The remaining half (8) were processed the following day to assess impact of holding/shipping time. A total of 2544 samples were analyzed including replicates. After accounting for inherent measurement error, we found that the effect of holding time on observed Legionella counts was small and should have no practical impact on interpretation of results. Holding samples increased the root mean squared error by only about 3-8%. Notably, for only one of 159 samples, did the average of the 8 replicate counts change by 1 log. Thus, our findings do not support the hypothesis of frequent, significant (≥= 1 log10 unit) Legionella colony count changes due to holding. Copyright © 2014 The Authors. Published by Elsevier Ltd.. All rights reserved.
Modified fast frequency acquisition via adaptive least squares algorithm
NASA Technical Reports Server (NTRS)
Kumar, Rajendra (Inventor)
1992-01-01
A method and the associated apparatus for estimating the amplitude, frequency, and phase of a signal of interest are presented. The method comprises the following steps: (1) inputting the signal of interest; (2) generating a reference signal with adjustable amplitude, frequency and phase at an output thereof; (3) mixing the signal of interest with the reference signal and a signal 90 deg out of phase with the reference signal to provide a pair of quadrature sample signals comprising respectively a difference between the signal of interest and the reference signal and a difference between the signal of interest and the signal 90 deg out of phase with the reference signal; (4) using the pair of quadrature sample signals to compute estimates of the amplitude, frequency, and phase of an error signal comprising the difference between the signal of interest and the reference signal employing a least squares estimation; (5) adjusting the amplitude, frequency, and phase of the reference signal from the numerically controlled oscillator in a manner which drives the error signal towards zero; and (6) outputting the estimates of the amplitude, frequency, and phase of the error signal in combination with the reference signal to produce a best estimate of the amplitude, frequency, and phase of the signal of interest. The preferred method includes the step of providing the error signal as a real time confidence measure as to the accuracy of the estimates wherein the closer the error signal is to zero, the higher the probability that the estimates are accurate. A matrix in the estimation algorithm provides an estimate of the variance of the estimation error.
Statistical modelling of thermal annealing of fission tracks in apatite
NASA Astrophysics Data System (ADS)
Laslett, G. M.; Galbraith, R. F.
1996-12-01
We develop an improved methodology for modelling the relationship between mean track length, temperature, and time in fission track annealing experiments. We consider "fanning Arrhenius" models, in which contours of constant mean length on an Arrhenius plot are straight lines meeting at a common point. Features of our approach are explicit use of subject matter knowledge, treating mean length as the response variable, modelling of the mean-variance relationship with two components of variance, improved modelling of the control sample, and using information from experiments in which no tracks are seen. This approach overcomes several weaknesses in previous models and provides a robust six parameter model that is widely applicable. Estimation is via direct maximum likelihood which can be implemented using a standard numerical optimisation package. Because the model is highly nonlinear, some reparameterisations are needed to achieve stable estimation and calculation of precisions. Experience suggests that precisions are more convincingly estimated from profile log-likelihood functions than from the information matrix. We apply our method to the B-5 and Sr fluorapatite data of Crowley et al. (1991) and obtain well-fitting models in both cases. For the B-5 fluorapatite, our model exhibits less fanning than that of Crowley et al. (1991), although fitted mean values above 12 μm are fairly similar. However, predictions can be different, particularly for heavy annealing at geological time scales, where our model is less retentive. In addition, the refined error structure of our model results in tighter prediction errors, and has components of error that are easier to verify or modify. For the Sr fluorapatite, our fitted model for mean lengths does not differ greatly from that of Crowley et al. (1991), but our error structure is quite different.
Natural variance in pH as a complication in detecting acidification of lakes
Turk, J.T.
1988-01-01
Natural variance in the pH of three dilute lakes in the Flat Tops Wilderness Area, Colorado, complicates the detection of acidification. Variations in pH during July-September of 1983 were: 0.95 (Ned Wilson Lake), 1.36 (Upper Island Lake), and 1.53 (Oyster Lake). Mean diurnal variations in pH during 1983 were: 0.37 (Ned Wilson Lake), 0.54 (Upper Island Lake), and 0.39 (Oyster Lake). Replicate pH measurements indicate that pH can be measured with a mean variance due to measurement error of ?? 0.005. Regression analysis indicates that samples collected on the same day of different years may differ because of time of day and percentage of cloud cover. Differences in wind duration and intensity and primary productivity also may cause the pH to differ between years. Such differences can be either random or systematic. Comparisons of pH among 3 yr of data from Ned Wilson Lake indicate that natural variations in pH are much larger than variations in Colorado Lakes previously attributed to acidification by precipitation.
EGSIEM combination service: combination of GRACE monthly K-band solutions on normal equation level
NASA Astrophysics Data System (ADS)
Meyer, Ulrich; Jean, Yoomin; Arnold, Daniel; Jäggi, Adrian
2017-04-01
The European Gravity Service for Improved Emergency Management (EGSIEM) project offers a scientific combination service, combining for the first time monthly GRACE gravity fields of different analysis centers (ACs) on normal equation (NEQ) level and thus taking all correlations between the gravity field coefficients and pre-eliminated orbit and instrument parameters correctly into account. Optimal weights for the individual NEQs are commonly derived by variance component estimation (VCE), as is the case for the products of the International VLBI Service (IVS) or the DTRF2008 reference frame realisation that are also derived by combination on NEQ-level. But variance factors are based on post-fit residuals and strongly depend on observation sampling and noise modeling, which both are very diverse in case of the individual EGSIEM ACs. These variance factors do not necessarily represent the true error levels of the estimated gravity field parameters that are still governed by analysis noise. We present a combination approach where weights are derived on solution level, thereby taking the analysis noise into account.
ERIC Educational Resources Information Center
Rodrigo, Ma. Mercedes T.; Andallaza, Thor Collin S.; Castro, Francisco Enrique Vicente G.; Armenta, Marc Lester V.; Dy, Thomas T.; Jadud, Matthew C.
2013-01-01
In this article we quantitatively and qualitatively analyze a sample of novice programmer compilation log data, exploring whether (or how) low-achieving, average, and high-achieving students vary in their grasp of these introductory concepts. High-achieving students self-reported having the easiest time learning the introductory programming…
Data Analysis and Its Impact on Predicting Schedule & Cost Risk
2006-03-01
variance of the error term by performing a Breusch - Pagan test for constant variance (Neter et al., 1996:239). In order to test the normality of...is constant variance. Using Microsoft Excel®, we calculate a p- 68 value of 0.225678 for the Breusch - Pagan test . We again compare this p-value to...calculate a p-value of 0.121211092 Breusch - Pagan test . We again compare this p-value to an alpha of 0.05 indicating our assumption of constant variance
Etzel, C J; Shete, S; Beasley, T M; Fernandez, J R; Allison, D B; Amos, C I
2003-01-01
Non-normality of the phenotypic distribution can affect power to detect quantitative trait loci in sib pair studies. Previously, we observed that Winsorizing the sib pair phenotypes increased the power of quantitative trait locus (QTL) detection for both Haseman-Elston (HE) least-squares tests [Hum Hered 2002;53:59-67] and maximum likelihood-based variance components (MLVC) analysis [Behav Genet (in press)]. Winsorizing the phenotypes led to a slight increase in type 1 error in H-E tests and a slight decrease in type I error for MLVC analysis. Herein, we considered transforming the sib pair phenotypes using the Box-Cox family of transformations. Data were simulated for normal and non-normal (skewed and kurtic) distributions. Phenotypic values were replaced by Box-Cox transformed values. Twenty thousand replications were performed for three H-E tests of linkage and the likelihood ratio test (LRT), the Wald test and other robust versions based on the MLVC method. We calculated the relative nominal inflation rate as the ratio of observed empirical type 1 error divided by the set alpha level (5, 1 and 0.1% alpha levels). MLVC tests applied to non-normal data had inflated type I errors (rate ratio greater than 1.0), which were controlled best by Box-Cox transformation and to a lesser degree by Winsorizing. For example, for non-transformed, skewed phenotypes (derived from a chi2 distribution with 2 degrees of freedom), the rates of empirical type 1 error with respect to set alpha level=0.01 were 0.80, 4.35 and 7.33 for the original H-E test, LRT and Wald test, respectively. For the same alpha level=0.01, these rates were 1.12, 3.095 and 4.088 after Winsorizing and 0.723, 1.195 and 1.905 after Box-Cox transformation. Winsorizing reduced inflated error rates for the leptokurtic distribution (derived from a Laplace distribution with mean 0 and variance 8). Further, power (adjusted for empirical type 1 error) at the 0.01 alpha level ranged from 4.7 to 17.3% across all tests using the non-transformed, skewed phenotypes, from 7.5 to 20.1% after Winsorizing and from 12.6 to 33.2% after Box-Cox transformation. Likewise, power (adjusted for empirical type 1 error) using leptokurtic phenotypes at the 0.01 alpha level ranged from 4.4 to 12.5% across all tests with no transformation, from 7 to 19.2% after Winsorizing and from 4.5 to 13.8% after Box-Cox transformation. Thus the Box-Cox transformation apparently provided the best type 1 error control and maximal power among the procedures we considered for analyzing a non-normal, skewed distribution (chi2) while Winzorizing worked best for the non-normal, kurtic distribution (Laplace). We repeated the same simulations using a larger sample size (200 sib pairs) and found similar results. Copyright 2003 S. Karger AG, Basel
Mathematical modeling of sample stacking methods in microfluidic systems
NASA Astrophysics Data System (ADS)
Horek, Jon
Gradient focusing methods are a general class of experimental techniques used to simultaneously separate and increase the cross-sectionally averaged concentration of charged particle mixtures. In comparison, Field Amplified Sample Stacking (FASS) techniques first concentrate the collection of molecules before separating them. Together, we denote gradient focusing and FASS methods "sample stacking" and study the dynamics of a specific method, Temperature Gradient Focusing (TGF), in which an axial temperature gradient is applied along a channel filled with weak buffer. Gradients in electroosmotic fluid flow and electrophoretic species velocity create the simultaneous separating and concentrating mechanism mentioned above. In this thesis, we begin with the observation that very little has been done to model the dynamics of gradient focusing, and proceed to solve the fundamental equations of fluid mechanics and scalar transport, assuming the existence of slow axial variations and the Taylor-Aris dispersion coefficient. In doing so, asymptotic methods reduce the equations from 3D to 1D, and we arrive at a simple 1D model which can be used to predict the transient evolution of the cross-sectionally averaged analyte concentration. In the second half of this thesis, we run several numerical focusing experiments with a 3D finite volume code. Comparison of the 1D theory and 3D simulations illustrates not only that the asymptotic theory converges as a certain parameter tends to zero, but also that fairly large axial slip velocity gradients lead to quite small errors in predicted steady variance. Additionally, we observe that the axial asymmetry of the electrophoretic velocity model leads to asymmetric peak shapes, a violation of the symmetric Gaussians predicted by the 1D theory. We conclude with some observations on the effect of Peclet number and gradient strength on the performance of focusing experiments, and describe a method for experimental optimization. Such knowledge is useful for design of lab-on-a-chip devices.
Bias correction for selecting the minimal-error classifier from many machine learning models.
Ding, Ying; Tang, Shaowu; Liao, Serena G; Jia, Jia; Oesterreich, Steffi; Lin, Yan; Tseng, George C
2014-11-15
Supervised machine learning is commonly applied in genomic research to construct a classifier from the training data that is generalizable to predict independent testing data. When test datasets are not available, cross-validation is commonly used to estimate the error rate. Many machine learning methods are available, and it is well known that no universally best method exists in general. It has been a common practice to apply many machine learning methods and report the method that produces the smallest cross-validation error rate. Theoretically, such a procedure produces a selection bias. Consequently, many clinical studies with moderate sample sizes (e.g. n = 30-60) risk reporting a falsely small cross-validation error rate that could not be validated later in independent cohorts. In this article, we illustrated the probabilistic framework of the problem and explored the statistical and asymptotic properties. We proposed a new bias correction method based on learning curve fitting by inverse power law (IPL) and compared it with three existing methods: nested cross-validation, weighted mean correction and Tibshirani-Tibshirani procedure. All methods were compared in simulation datasets, five moderate size real datasets and two large breast cancer datasets. The result showed that IPL outperforms the other methods in bias correction with smaller variance, and it has an additional advantage to extrapolate error estimates for larger sample sizes, a practical feature to recommend whether more samples should be recruited to improve the classifier and accuracy. An R package 'MLbias' and all source files are publicly available. tsenglab.biostat.pitt.edu/software.htm. ctseng@pitt.edu Supplementary data are available at Bioinformatics online. © The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.
Measuring kinetics of complex single ion channel data using mean-variance histograms.
Patlak, J B
1993-01-01
The measurement of single ion channel kinetics is difficult when those channels exhibit subconductance events. When the kinetics are fast, and when the current magnitudes are small, as is the case for Na+, Ca2+, and some K+ channels, these difficulties can lead to serious errors in the estimation of channel kinetics. I present here a method, based on the construction and analysis of mean-variance histograms, that can overcome these problems. A mean-variance histogram is constructed by calculating the mean current and the current variance within a brief "window" (a set of N consecutive data samples) superimposed on the digitized raw channel data. Systematic movement of this window over the data produces large numbers of mean-variance pairs which can be assembled into a two-dimensional histogram. Defined current levels (open, closed, or sublevel) appear in such plots as low variance regions. The total number of events in such low variance regions is estimated by curve fitting and plotted as a function of window width. This function decreases with the same time constants as the original dwell time probability distribution for each of the regions. The method can therefore be used: 1) to present a qualitative summary of the single channel data from which the signal-to-noise ratio, open channel noise, steadiness of the baseline, and number of conductance levels can be quickly determined; 2) to quantify the dwell time distribution in each of the levels exhibited. In this paper I present the analysis of a Na+ channel recording that had a number of complexities. The signal-to-noise ratio was only about 8 for the main open state, open channel noise, and fast flickers to other states were present, as were a substantial number of subconductance states. "Standard" half-amplitude threshold analysis of these data produce open and closed time histograms that were well fitted by the sum of two exponentials, but with apparently erroneous time constants, whereas the mean-variance histogram technique provided a more credible analysis of the open, closed, and subconductance times for the patch. I also show that the method produces accurate results on simulated data in a wide variety of conditions, whereas the half-amplitude method, when applied to complex simulated data shows the same errors as were apparent in the real data. The utility and the limitations of this new method are discussed. Images FIGURE 2 FIGURE 4 FIGURE 8 FIGURE 9 PMID:7690261
NASA Astrophysics Data System (ADS)
Li, Xiongwei; Wang, Zhe; Lui, Siu-Lung; Fu, Yangting; Li, Zheng; Liu, Jianming; Ni, Weidou
2013-10-01
A bottleneck of the wide commercial application of laser-induced breakdown spectroscopy (LIBS) technology is its relatively high measurement uncertainty. A partial least squares (PLS) based normalization method was proposed to improve pulse-to-pulse measurement precision for LIBS based on our previous spectrum standardization method. The proposed model utilized multi-line spectral information of the measured element and characterized the signal fluctuations due to the variation of plasma characteristic parameters (plasma temperature, electron number density, and total number density) for signal uncertainty reduction. The model was validated by the application of copper concentration prediction in 29 brass alloy samples. The results demonstrated an improvement on both measurement precision and accuracy over the generally applied normalization as well as our previously proposed simplified spectrum standardization method. The average relative standard deviation (RSD), average of the standard error (error bar), the coefficient of determination (R2), the root-mean-square error of prediction (RMSEP), and average value of the maximum relative error (MRE) were 1.80%, 0.23%, 0.992, 1.30%, and 5.23%, respectively, while those for the generally applied spectral area normalization were 3.72%, 0.71%, 0.973, 1.98%, and 14.92%, respectively.
Siegert, Wolfgang; Ganzer, Christian; Kluth, Holger; Rodehutscord, Markus
2018-06-01
A regression approach was applied to determine the influence of feed provisioning prior to digesta sampling on precaecal (pc) amino acid (AA) digestibility in broiler chickens. Soybean meal was used as an example test ingredient. Five feed-provisioning protocols were investigated, four with restricted provision and one with ad libitum provision. When provision was restricted, feed was provided for 30 min after a withdrawal period of 12 h. Digesta were sampled 1, 2, 4 and 6 h after feeding commenced. A diet containing 300 g maize starch/kg was prepared. Half or all the maize starch was replaced with soybean meal in two other diets. Average pc digestibility of all determined AA in the soybean meal was 86% for the 4 and 6-h protocols and 66% and 60% for the 2 and 1-h protocols, respectively. Average pc AA digestibility of soybean meal was 76% for ad libitum feed provision. Feed provisioning also influenced the determined variance. Variance in digestibility ranked in magnitude 1 h > ad libitum > 2 h > 6 h > 4 h for all AA. Owing to the considerable influence of feed-provisioning protocols found in this study, comparisons of pc AA digestibility between studies applying different protocols prior to digesta sampling must be treated with caution. Digestibility experiments aimed at providing estimates for practical feed formulation should use feed-provisioning procedures similar to those used in practice.
Olives, Casey; Pagano, Marcello; Deitchler, Megan; Hedt, Bethany L; Egge, Kari; Valadez, Joseph J
2009-04-01
Traditional lot quality assurance sampling (LQAS) methods require simple random sampling to guarantee valid results. However, cluster sampling has been proposed to reduce the number of random starting points. This study uses simulations to examine the classification error of two such designs, a 67x3 (67 clusters of three observations) and a 33x6 (33 clusters of six observations) sampling scheme to assess the prevalence of global acute malnutrition (GAM). Further, we explore the use of a 67x3 sequential sampling scheme for LQAS classification of GAM prevalence. Results indicate that, for independent clusters with moderate intracluster correlation for the GAM outcome, the three sampling designs maintain approximate validity for LQAS analysis. Sequential sampling can substantially reduce the average sample size that is required for data collection. The presence of intercluster correlation can impact dramatically the classification error that is associated with LQAS analysis.
Bøcher, Peder Klith; McCloy, Keith R
2006-02-01
In this investigation, the characteristics of the average local variance (ALV) function is investigated through the acquisition of images at different spatial resolutions of constructed scenes of regular patterns of black and white squares. It is shown that the ALV plot consistently peaks at a spatial resolution in which the pixels has a size corresponding to half the distance between scene objects, and that, under very specific conditions, it also peaks at a spatial resolution in which the pixel size corresponds to the whole distance between scene objects. It is argued that the peak at object distance when present is an expression of the Nyquist sample rate. The presence of this peak is, hence, shown to be a function of the matching between the phase of the scene pattern and the phase of the sample grid, i.e., the image. When these phases match, a clear and distinct peak is produced on the ALV plot. The fact that the peak at half the distance consistently occurs in the ALV plot is linked to the circumstance that the sampling interval (distance between pixels) and the extent of the sampling unit (size of pixels) are equal. Hence, at twice the Nyquist sampling rate, each fundamental period of the pattern is covered by four pixels; therefore, at least one pixel is always completely embedded within one pattern element, regardless of sample scene phase. If the objects in the scene are scattered with a distance larger than their extent, the peak will be related to the size by a factor larger than 1/2. This is suggested to be the explanation to the results presented by others that the ALV plot is related to scene-object size by a factor of 1/2-3/4.
NASA Astrophysics Data System (ADS)
Wan, Tat C.; Kabuka, Mansur R.
1994-05-01
With the tremendous growth in imaging applications and the development of filmless radiology, the need for compression techniques that can achieve high compression ratios with user specified distortion rates becomes necessary. Boundaries and edges in the tissue structures are vital for detection of lesions and tumors, which in turn requires the preservation of edges in the image. The proposed edge preserving image compressor (EPIC) combines lossless compression of edges with neural network compression techniques based on dynamic associative neural networks (DANN), to provide high compression ratios with user specified distortion rates in an adaptive compression system well-suited to parallel implementations. Improvements to DANN-based training through the use of a variance classifier for controlling a bank of neural networks speed convergence and allow the use of higher compression ratios for `simple' patterns. The adaptation and generalization capabilities inherent in EPIC also facilitate progressive transmission of images through varying the number of quantization levels used to represent compressed patterns. Average compression ratios of 7.51:1 with an averaged average mean squared error of 0.0147 were achieved.
Mineral composition of Atriplex hymenelytra growing in the northern Mojave Desert
DOE Office of Scientific and Technical Information (OSTI.GOV)
Wallace, A.; Romney, E.M.; Hunter, R.B.
1980-01-01
Fifty samples of Atriplex hymenelytra (Torr.) S. Wats. were collected from several different locations in southern Nevada and California to test variability in mineral composition. Only Na, V, P, Ca, Mg, Mn, and Sr in the samples appeared to represent a uniform population resulting in normal curves for frequency distribution. Even so, about 40 percent of the variance for these elements was due to location. All elements differed enough with location so that no element really represented a uniform population. The coefficient of variation for most elements was over 40 percent and one was over 100 percent. The proportion ofmore » variance due to analytical variation averaged 16.2 +- 13.1 percent (standard deviation), that due to location was 43.0 +- 13.4 percent, and that due to variation of plants within location was 40.7 +- 13.0 percent.« less
Two-sample tests and one-way MANOVA for multivariate biomarker data with nondetects.
Thulin, M
2016-09-10
Testing whether the mean vector of a multivariate set of biomarkers differs between several populations is an increasingly common problem in medical research. Biomarker data is often left censored because some measurements fall below the laboratory's detection limit. We investigate how such censoring affects multivariate two-sample and one-way multivariate analysis of variance tests. Type I error rates, power and robustness to increasing censoring are studied, under both normality and non-normality. Parametric tests are found to perform better than non-parametric alternatives, indicating that the current recommendations for analysis of censored multivariate data may have to be revised. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.
A Posteriori Correction of Forecast and Observation Error Variances
NASA Technical Reports Server (NTRS)
Rukhovets, Leonid
2005-01-01
Proposed method of total observation and forecast error variance correction is based on the assumption about normal distribution of "observed-minus-forecast" residuals (O-F), where O is an observed value and F is usually a short-term model forecast. This assumption can be accepted for several types of observations (except humidity) which are not grossly in error. Degree of nearness to normal distribution can be estimated by the symmetry or skewness (luck of symmetry) a(sub 3) = mu(sub 3)/sigma(sup 3) and kurtosis a(sub 4) = mu(sub 4)/sigma(sup 4) - 3 Here mu(sub i) = i-order moment, sigma is a standard deviation. It is well known that for normal distribution a(sub 3) = a(sub 4) = 0.
Growth models and the expected distribution of fluctuating asymmetry
Graham, John H.; Shimizu, Kunio; Emlen, John M.; Freeman, D. Carl; Merkel, John
2003-01-01
Multiplicative error accounts for much of the size-scaling and leptokurtosis in fluctuating asymmetry. It arises when growth involves the addition of tissue to that which is already present. Such errors are lognormally distributed. The distribution of the difference between two lognormal variates is leptokurtic. If those two variates are correlated, then the asymmetry variance will scale with size. Inert tissues typically exhibit additive error and have a gamma distribution. Although their asymmetry variance does not exhibit size-scaling, the distribution of the difference between two gamma variates is nevertheless leptokurtic. Measurement error is also additive, but has a normal distribution. Thus, the measurement of fluctuating asymmetry may involve the mixing of additive and multiplicative error. When errors are multiplicative, we recommend computing log E(l) − log E(r), the difference between the logarithms of the expected values of left and right sides, even when size-scaling is not obvious. If l and r are lognormally distributed, and measurement error is nil, the resulting distribution will be normal, and multiplicative error will not confound size-related changes in asymmetry. When errors are additive, such a transformation to remove size-scaling is unnecessary. Nevertheless, the distribution of l − r may still be leptokurtic.
USDA-ARS?s Scientific Manuscript database
If not properly account for, auto-correlated errors in observations can lead to inaccurate results in soil moisture data analysis and reanalysis. Here, we propose a more generalized form of the triple collocation algorithm (GTC) capable of decomposing the total error variance of remotely-sensed surf...
Training set optimization under population structure in genomic selection.
Isidro, Julio; Jannink, Jean-Luc; Akdemir, Deniz; Poland, Jesse; Heslot, Nicolas; Sorrells, Mark E
2015-01-01
Population structure must be evaluated before optimization of the training set population. Maximizing the phenotypic variance captured by the training set is important for optimal performance. The optimization of the training set (TRS) in genomic selection has received much interest in both animal and plant breeding, because it is critical to the accuracy of the prediction models. In this study, five different TRS sampling algorithms, stratified sampling, mean of the coefficient of determination (CDmean), mean of predictor error variance (PEVmean), stratified CDmean (StratCDmean) and random sampling, were evaluated for prediction accuracy in the presence of different levels of population structure. In the presence of population structure, the most phenotypic variation captured by a sampling method in the TRS is desirable. The wheat dataset showed mild population structure, and CDmean and stratified CDmean methods showed the highest accuracies for all the traits except for test weight and heading date. The rice dataset had strong population structure and the approach based on stratified sampling showed the highest accuracies for all traits. In general, CDmean minimized the relationship between genotypes in the TRS, maximizing the relationship between TRS and the test set. This makes it suitable as an optimization criterion for long-term selection. Our results indicated that the best selection criterion used to optimize the TRS seems to depend on the interaction of trait architecture and population structure.
Robust LOD scores for variance component-based linkage analysis.
Blangero, J; Williams, J T; Almasy, L
2000-01-01
The variance component method is now widely used for linkage analysis of quantitative traits. Although this approach offers many advantages, the importance of the underlying assumption of multivariate normality of the trait distribution within pedigrees has not been studied extensively. Simulation studies have shown that traits with leptokurtic distributions yield linkage test statistics that exhibit excessive Type I error when analyzed naively. We derive analytical formulae relating the deviation from the expected asymptotic distribution of the lod score to the kurtosis and total heritability of the quantitative trait. A simple correction constant yields a robust lod score for any deviation from normality and for any pedigree structure, and effectively eliminates the problem of inflated Type I error due to misspecification of the underlying probability model in variance component-based linkage analysis.
Spectral Analysis of Forecast Error Investigated with an Observing System Simulation Experiment
NASA Technical Reports Server (NTRS)
Prive, N. C.; Errico, Ronald M.
2015-01-01
The spectra of analysis and forecast error are examined using the observing system simulation experiment (OSSE) framework developed at the National Aeronautics and Space Administration Global Modeling and Assimilation Office (NASAGMAO). A global numerical weather prediction model, the Global Earth Observing System version 5 (GEOS-5) with Gridpoint Statistical Interpolation (GSI) data assimilation, is cycled for two months with once-daily forecasts to 336 hours to generate a control case. Verification of forecast errors using the Nature Run as truth is compared with verification of forecast errors using self-analysis; significant underestimation of forecast errors is seen using self-analysis verification for up to 48 hours. Likewise, self analysis verification significantly overestimates the error growth rates of the early forecast, as well as mischaracterizing the spatial scales at which the strongest growth occurs. The Nature Run-verified error variances exhibit a complicated progression of growth, particularly for low wave number errors. In a second experiment, cycling of the model and data assimilation over the same period is repeated, but using synthetic observations with different explicitly added observation errors having the same error variances as the control experiment, thus creating a different realization of the control. The forecast errors of the two experiments become more correlated during the early forecast period, with correlations increasing for up to 72 hours before beginning to decrease.
Rate, causes and reporting of medication errors in Jordan: nurses' perspectives.
Mrayyan, Majd T; Shishani, Kawkab; Al-Faouri, Ibrahim
2007-09-01
The aim of the study was to describe Jordanian nurses' perceptions about various issues related to medication errors. This is the first nursing study about medication errors in Jordan. This was a descriptive study. A convenient sample of 799 nurses from 24 hospitals was obtained. Descriptive and inferential statistics were used for data analysis. Over the course of their nursing career, the average number of recalled committed medication errors per nurse was 2.2. Using incident reports, the rate of medication errors reported to nurse managers was 42.1%. Medication errors occurred mainly when medication labels/packaging were of poor quality or damaged. Nurses failed to report medication errors because they were afraid that they might be subjected to disciplinary actions or even lose their jobs. In the stepwise regression model, gender was the only predictor of medication errors in Jordan. Strategies to reduce or eliminate medication errors are required.
Essays in financial economics and econometrics
NASA Astrophysics Data System (ADS)
La Spada, Gabriele
Chapter 1 (my job market paper) asks the following question: Do asset managers reach for yield because of competitive pressures in a low rate environment? I propose a tournament model of money market funds (MMFs) to study this issue. I show that funds with different costs of default respond differently to changes in interest rates, and that it is important to distinguish the role of risk-free rates from that of risk premia. An increase in the risk premium leads funds with lower default costs to increase risk-taking, while funds with higher default costs reduce risk-taking. Without changes in the premium, low risk-free rates reduce risk-taking. My empirical analysis shows that these predictions are consistent with the risk-taking of MMFs during the 2006--2008 period. Chapter 2, co-authored with Fabrizio Lillo and published in Studies in Nonlinear Dynamics and Econometrics (2014), studies the effect of round-off error (or discretization) on stationary Gaussian long-memory process. For large lags, the autocovariance is rescaled by a factor smaller than one, and we compute this factor exactly. Hence, the discretized process has the same Hurst exponent as the underlying one. We show that in presence of round-off error, two common estimators of the Hurst exponent, the local Whittle (LW) estimator and the detrended fluctuation analysis (DFA), are severely negatively biased in finite samples. We derive conditions for consistency and asymptotic normality of the LW estimator applied to discretized processes and compute the asymptotic properties of the DFA for generic long-memory processes that encompass discretized processes. Chapter 3, co-authored with Fabrizio Lillo, studies the effect of round-off error on integrated Gaussian processes with possibly correlated increments. We derive the variance and kurtosis of the realized increment process in the limit of both "small" and "large" round-off errors, and its autocovariance for large lags. We propose novel estimators for the variance and lag-one autocorrelation of the underlying, unobserved increment process. We also show that for fractionally integrated processes, the realized increments have the same Hurst exponent as the underlying ones, but the LW estimator applied to the realized series is severely negatively biased in medium-sized samples.
NASA Astrophysics Data System (ADS)
Atta, Abdu; Yahaya, Sharipah; Zain, Zakiyah; Ahmed, Zalikha
2017-11-01
Control chart is established as one of the most powerful tools in Statistical Process Control (SPC) and is widely used in industries. The conventional control charts rely on normality assumption, which is not always the case for industrial data. This paper proposes a new S control chart for monitoring process dispersion using skewness correction method for skewed distributions, named as SC-S control chart. Its performance in terms of false alarm rate is compared with various existing control charts for monitoring process dispersion, such as scaled weighted variance S chart (SWV-S); skewness correction R chart (SC-R); weighted variance R chart (WV-R); weighted variance S chart (WV-S); and standard S chart (STD-S). Comparison with exact S control chart with regards to the probability of out-of-control detections is also accomplished. The Weibull and gamma distributions adopted in this study are assessed along with the normal distribution. Simulation study shows that the proposed SC-S control chart provides good performance of in-control probabilities (Type I error) in almost all the skewness levels and sample sizes, n. In the case of probability of detection shift the proposed SC-S chart is closer to the exact S control chart than the existing charts for skewed distributions, except for the SC-R control chart. In general, the performance of the proposed SC-S control chart is better than all the existing control charts for monitoring process dispersion in the cases of Type I error and probability of detection shift.
NASA Technical Reports Server (NTRS)
Li, Zhijin; Chao, Yi; McWilliams, James C.; Ide, Kayo
2008-01-01
A three-dimensional variational data assimilation scheme for the Regional Ocean Modeling System (ROMS), named ROMS3DVAR, has been described in the work of Li et al. (2008). In this paper, ROMS3DVAR is applied to the central California coastal region, an area characterized by inhomogeneity and anisotropy, as well as by dynamically unbalanced flows. A method for estimating the model error variances from limited observations is presented, and the construction of the inhomogeneous and anisotropic error correlations based on the Kronecker product is demonstrated. A set of single observation experiments illustrates the inhomogeneous and anisotropic error correlations and weak dynamic constraints used. Results are presented from the assimilation of data gathered during the Autonomous Ocean Sampling Network (AOSN) experiment during August 2003. The results show that ROMS3DVAR is capable of reproducing complex flows associated with upwelling and relaxation, as well as the rapid transitions between them. Some difficulties encountered during the experiment are also discussed.
Tarrab, Leticia; Garcia, Carlos M.; Cantero, Mariano I.; Oberg, Kevin
2012-01-01
This work presents a systematic analysis quantifying the role of the presence of turbulence fluctuations on uncertainties (random errors) of acoustic Doppler current profiler (ADCP) discharge measurements from moving platforms. Data sets of three-dimensional flow velocities with high temporal and spatial resolution were generated from direct numerical simulation (DNS) of turbulent open channel flow. Dimensionless functions relating parameters quantifying the uncertainty in discharge measurements due to flow turbulence (relative variance and relative maximum random error) to sampling configuration were developed from the DNS simulations and then validated with field-scale discharge measurements. The validated functions were used to evaluate the role of the presence of flow turbulence fluctuations on uncertainties in ADCP discharge measurements. The results of this work indicate that random errors due to the flow turbulence are significant when: (a) a low number of transects is used for a discharge measurement, and (b) measurements are made in shallow rivers using high boat velocity (short time for the boat to cross a flow turbulence structure).
Hand-Writing Motion Tracking with Vision-Inertial Sensor Fusion: Calibration and Error Correction
Zhou, Shengli; Fei, Fei; Zhang, Guanglie; Liu, Yunhui; Li, Wen J.
2014-01-01
The purpose of this study was to improve the accuracy of real-time ego-motion tracking through inertial sensor and vision sensor fusion. Due to low sampling rates supported by web-based vision sensor and accumulation of errors in inertial sensors, ego-motion tracking with vision sensors is commonly afflicted by slow updating rates, while motion tracking with inertial sensor suffers from rapid deterioration in accuracy with time. This paper starts with a discussion of developed algorithms for calibrating two relative rotations of the system using only one reference image. Next, stochastic noises associated with the inertial sensor are identified using Allan Variance analysis, and modeled according to their characteristics. Finally, the proposed models are incorporated into an extended Kalman filter for inertial sensor and vision sensor fusion. Compared with results from conventional sensor fusion models, we have shown that ego-motion tracking can be greatly enhanced using the proposed error correction model. PMID:25157546
The validity of two clinical tests of visual-motor perception.
Wallbrown, J D; Wallbrown, F H; Engin, A W
1977-04-01
The study investigated the relative efficiency of the Bender and MPD as assessors of achievement-related errors in visual-motor perception. Clinical experience with these two tests suggests that beyond first grade the MPD is more sensitive than the Bender for purposes of measuring deficits in visual-motor perception that interfere with effective classroom learning. The sample was composed of 153 third-grade children from two upper-middle-class elementary schools in a surburban school system in central Ohio. For three of the four achievement criteria, the results were clearly congruent with the hypothesis stated above. That is, SpCD errors from the MPD not only showed significantly higher negative rs with the criteria (reading vocabulary, reading comprehension, and mathematics computation) than Koppitz errors from the Bender, but also accounted for a much higher proportion of the variance in these criteria. Thus, the findings suggest that psychologists engaged in the assessment of older children seriously should consider adding the MPD to their assessment battery.
The impact of multiple endpoint dependency on Q and I(2) in meta-analysis.
Thompson, Christopher Glen; Becker, Betsy Jane
2014-09-01
A common assumption in meta-analysis is that effect sizes are independent. When correlated effect sizes are analyzed using traditional univariate techniques, this assumption is violated. This research assesses the impact of dependence arising from treatment-control studies with multiple endpoints on homogeneity measures Q and I(2) in scenarios using the unbiased standardized-mean-difference effect size. Univariate and multivariate meta-analysis methods are examined. Conditions included different overall outcome effects, study sample sizes, numbers of studies, between-outcomes correlations, dependency structures, and ways of computing the correlation. The univariate approach used typical fixed-effects analyses whereas the multivariate approach used generalized least-squares (GLS) estimates of a fixed-effects model, weighted by the inverse variance-covariance matrix. Increased dependence among effect sizes led to increased Type I error rates from univariate models. When effect sizes were strongly dependent, error rates were drastically higher than nominal levels regardless of study sample size and number of studies. In contrast, using GLS estimation to account for multiple-endpoint dependency maintained error rates within nominal levels. Conversely, mean I(2) values were not greatly affected by increased amounts of dependency. Last, we point out that the between-outcomes correlation should be estimated as a pooled within-groups correlation rather than using a full-sample estimator that does not consider treatment/control group membership. Copyright © 2014 John Wiley & Sons, Ltd.
Statistics of the epoch of reionization 21-cm signal - I. Power spectrum error-covariance
NASA Astrophysics Data System (ADS)
Mondal, Rajesh; Bharadwaj, Somnath; Majumdar, Suman
2016-02-01
The non-Gaussian nature of the epoch of reionization (EoR) 21-cm signal has a significant impact on the error variance of its power spectrum P(k). We have used a large ensemble of seminumerical simulations and an analytical model to estimate the effect of this non-Gaussianity on the entire error-covariance matrix {C}ij. Our analytical model shows that {C}ij has contributions from two sources. One is the usual variance for a Gaussian random field which scales inversely of the number of modes that goes into the estimation of P(k). The other is the trispectrum of the signal. Using the simulated 21-cm Signal Ensemble, an ensemble of the Randomized Signal and Ensembles of Gaussian Random Ensembles we have quantified the effect of the trispectrum on the error variance {C}II. We find that its relative contribution is comparable to or larger than that of the Gaussian term for the k range 0.3 ≤ k ≤ 1.0 Mpc-1, and can be even ˜200 times larger at k ˜ 5 Mpc-1. We also establish that the off-diagonal terms of {C}ij have statistically significant non-zero values which arise purely from the trispectrum. This further signifies that the error in different k modes are not independent. We find a strong correlation between the errors at large k values (≥0.5 Mpc-1), and a weak correlation between the smallest and largest k values. There is also a small anticorrelation between the errors in the smallest and intermediate k values. These results are relevant for the k range that will be probed by the current and upcoming EoR 21-cm experiments.
Differences in head impulse test results due to analysis techniques.
Cleworth, Taylor W; Carpenter, Mark G; Honegger, Flurin; Allum, John H J
2017-01-01
Different analysis techniques are used to define vestibulo-ocular reflex (VOR) gain between eye and head angular velocity during the video head impulse test (vHIT). Comparisons would aid selection of gain techniques best related to head impulse characteristics and promote standardisation. Compare and contrast known methods of calculating vHIT VOR gain. We examined lateral canal vHIT responses recorded from 20 patients twice within 13 weeks of acute unilateral peripheral vestibular deficit onset. Ten patients were tested with an ICS Impulse system (GN Otometrics) and 10 with an EyeSeeCam (ESC) system (Interacoustics). Mean gain and variance were computed with area, average sample gain, and regression techniques over specific head angular velocity (HV) and acceleration (HA) intervals. Results for the same gain technique were not different between measurement systems. Area and average sample gain yielded equally lower variances than regression techniques. Gains computed over the whole impulse duration were larger than those computed for increasing HV. Gain over decreasing HV was associated with larger variances. Gains computed around peak HV were smaller than those computed around peak HA. The median gain over 50-70 ms was not different from gain around peak HV. However, depending on technique used, the gain over increasing HV was different from gain around peak HA. Conversion equations between gains obtained with standard ICS and ESC methods were computed. For low gains, the conversion was dominated by a constant that needed to be added to ESC gains to equal ICS gains. We recommend manufacturers standardize vHIT gain calculations using 2 techniques: area gain around peak HA and peak HV.
Effect of camera angulation on adaptation of CAD/CAM restorations.
Parsell, D E; Anderson, B C; Livingston, H M; Rudd, J I; Tankersley, J D
2000-01-01
A significant concern with computer-assisted design/computer-assisted manufacturing (CAD/CAM)-produced prostheses is the accuracy of adaptation of the restoration to the preparation. The objective of this study is to determine the effect of operator-controlled camera misalignment on restoration adaptation. A CEREC 2 CAD/CAM unit (Sirona Dental Systems, Bensheim, Germany) was used to capture the optical impressions and machine the restorations. A Class I preparation was used as the standard preparation for optical impressions. Camera angles along the mesio-distal and buccolingual alignment were varied from the ideal orientation. Occlusal marginal gaps and sample height, width, and length were measured and compared to preparation dimensions. For clinical correlation, clinicians were asked to take optical impressions of mesio-occlusal preparations (Class II) on all four second molar sites, using a patient simulator. On the adjacent first molar occlusal surfaces, a preparation was machined such that camera angulation could be calculated from information taken from the optical impression. Degree of tilt and plane of tilt were compared to the optimum camera positions for those preparations. One-way analysis of variance and Dunnett C post hoc testing (alpha = 0.01) revealed little significant degradation in fit with camera angulation. Only the apical length fit was significantly degraded by excessive angulation. The CEREC 2 CAD/CAM system was found to be relatively insensitive to operator-induced errors attributable to camera misalignments of less than 5 degrees in either the buccolingual or the mesiodistal plane. The average camera tilt error generated by clinicians for all sites was 1.98 +/- 1.17 degrees.