Intelligent Ensemble Forecasting System of Stock Market Fluctuations Based on Symetric and Asymetric Wavelet Functions

NASA Astrophysics Data System (ADS)

Lahmiri, Salim; Boukadoum, Mounir

2015-08-01

We present a new ensemble system for stock market returns prediction where continuous wavelet transform (CWT) is used to analyze return series and backpropagation neural networks (BPNNs) for processing CWT-based coefficients, determining the optimal ensemble weights, and providing final forecasts. Particle swarm optimization (PSO) is used for finding optimal weights and biases for each BPNN. To capture symmetry/asymmetry in the underlying data, three wavelet functions with different shapes are adopted. The proposed ensemble system was tested on three Asian stock markets: The Hang Seng, KOSPI, and Taiwan stock market data. Three statistical metrics were used to evaluate the forecasting accuracy; including, mean of absolute errors (MAE), root mean of squared errors (RMSE), and mean of absolute deviations (MADs). Experimental results showed that our proposed ensemble system outperformed the individual CWT-ANN models each with different wavelet function. In addition, the proposed ensemble system outperformed the conventional autoregressive moving average process. As a result, the proposed ensemble system is suitable to capture symmetry/asymmetry in financial data fluctuations for better prediction accuracy.

Unveiling Inherent Degeneracies in Determining Population-weighted Ensembles of Inter-domain Orientational Distributions Using NMR Residual Dipolar Couplings: Application to RNA Helix Junction Helix Motifs

PubMed Central

Yang, Shan; Al-Hashimi, Hashim M.

2016-01-01

A growing number of studies employ time-averaged experimental data to determine dynamic ensembles of biomolecules. While it is well known that different ensembles can satisfy experimental data to within error, the extent and nature of these degeneracies, and their impact on the accuracy of the ensemble determination remains poorly understood. Here, we use simulations and a recently introduced metric for assessing ensemble similarity to explore degeneracies in determining ensembles using NMR residual dipolar couplings (RDCs) with specific application to A-form helices in RNA. Various target ensembles were constructed representing different domain-domain orientational distributions that are confined to a topologically restricted (<10%) conformational space. Five independent sets of ensemble averaged RDCs were then computed for each target ensemble and a ‘sample and select’ scheme used to identify degenerate ensembles that satisfy RDCs to within experimental uncertainty. We find that ensembles with different ensemble sizes and that can differ significantly from the target ensemble (by as much as ΣΩ ~ 0.4 where ΣΩ varies between 0 and 1 for maximum and minimum ensemble similarity, respectively) can satisfy the ensemble averaged RDCs. These deviations increase with the number of unique conformers and breadth of the target distribution, and result in significant uncertainty in determining conformational entropy (as large as 5 kcal/mol at T = 298 K). Nevertheless, the RDC-degenerate ensembles are biased towards populated regions of the target ensemble, and capture other essential features of the distribution, including the shape. Our results identify ensemble size as a major source of uncertainty in determining ensembles and suggest that NMR interactions such as RDCs and spin relaxation, on their own, do not carry the necessary information needed to determine conformational entropy at a useful level of precision. The framework introduced here provides a general approach for exploring degeneracies in ensemble determination for different types of experimental data. PMID:26131693

MSEBAG: a dynamic classifier ensemble generation based on `minimum-sufficient ensemble' and bagging

NASA Astrophysics Data System (ADS)

Chen, Lei; Kamel, Mohamed S.

2016-01-01

In this paper, we propose a dynamic classifier system, MSEBAG, which is characterised by searching for the 'minimum-sufficient ensemble' and bagging at the ensemble level. It adopts an 'over-generation and selection' strategy and aims to achieve a good bias-variance trade-off. In the training phase, MSEBAG first searches for the 'minimum-sufficient ensemble', which maximises the in-sample fitness with the minimal number of base classifiers. Then, starting from the 'minimum-sufficient ensemble', a backward stepwise algorithm is employed to generate a collection of ensembles. The objective is to create a collection of ensembles with a descending fitness on the data, as well as a descending complexity in the structure. MSEBAG dynamically selects the ensembles from the collection for the decision aggregation. The extended adaptive aggregation (EAA) approach, a bagging-style algorithm performed at the ensemble level, is employed for this task. EAA searches for the competent ensembles using a score function, which takes into consideration both the in-sample fitness and the confidence of the statistical inference, and averages the decisions of the selected ensembles to label the test pattern. The experimental results show that the proposed MSEBAG outperforms the benchmarks on average.

Multi-model ensemble hydrologic prediction using Bayesian model averaging

NASA Astrophysics Data System (ADS)

Duan, Qingyun; Ajami, Newsha K.; Gao, Xiaogang; Sorooshian, Soroosh

2007-05-01

Multi-model ensemble strategy is a means to exploit the diversity of skillful predictions from different models. This paper studies the use of Bayesian model averaging (BMA) scheme to develop more skillful and reliable probabilistic hydrologic predictions from multiple competing predictions made by several hydrologic models. BMA is a statistical procedure that infers consensus predictions by weighing individual predictions based on their probabilistic likelihood measures, with the better performing predictions receiving higher weights than the worse performing ones. Furthermore, BMA provides a more reliable description of the total predictive uncertainty than the original ensemble, leading to a sharper and better calibrated probability density function (PDF) for the probabilistic predictions. In this study, a nine-member ensemble of hydrologic predictions was used to test and evaluate the BMA scheme. This ensemble was generated by calibrating three different hydrologic models using three distinct objective functions. These objective functions were chosen in a way that forces the models to capture certain aspects of the hydrograph well (e.g., peaks, mid-flows and low flows). Two sets of numerical experiments were carried out on three test basins in the US to explore the best way of using the BMA scheme. In the first set, a single set of BMA weights was computed to obtain BMA predictions, while the second set employed multiple sets of weights, with distinct sets corresponding to different flow intervals. In both sets, the streamflow values were transformed using Box-Cox transformation to ensure that the probability distribution of the prediction errors is approximately Gaussian. A split sample approach was used to obtain and validate the BMA predictions. The test results showed that BMA scheme has the advantage of generating more skillful and equally reliable probabilistic predictions than original ensemble. The performance of the expected BMA predictions in terms of daily root mean square error (DRMS) and daily absolute mean error (DABS) is generally superior to that of the best individual predictions. Furthermore, the BMA predictions employing multiple sets of weights are generally better than those using single set of weights.

Ensemble Methods for Classification of Physical Activities from Wrist Accelerometry.

PubMed

Chowdhury, Alok Kumar; Tjondronegoro, Dian; Chandran, Vinod; Trost, Stewart G

2017-09-01

To investigate whether the use of ensemble learning algorithms improve physical activity recognition accuracy compared to the single classifier algorithms, and to compare the classification accuracy achieved by three conventional ensemble machine learning methods (bagging, boosting, random forest) and a custom ensemble model comprising four algorithms commonly used for activity recognition (binary decision tree, k nearest neighbor, support vector machine, and neural network). The study used three independent data sets that included wrist-worn accelerometer data. For each data set, a four-step classification framework consisting of data preprocessing, feature extraction, normalization and feature selection, and classifier training and testing was implemented. For the custom ensemble, decisions from the single classifiers were aggregated using three decision fusion methods: weighted majority vote, naïve Bayes combination, and behavior knowledge space combination. Classifiers were cross-validated using leave-one subject out cross-validation and compared on the basis of average F1 scores. In all three data sets, ensemble learning methods consistently outperformed the individual classifiers. Among the conventional ensemble methods, random forest models provided consistently high activity recognition; however, the custom ensemble model using weighted majority voting demonstrated the highest classification accuracy in two of the three data sets. Combining multiple individual classifiers using conventional or custom ensemble learning methods can improve activity recognition accuracy from wrist-worn accelerometer data.

A model ensemble for projecting multi‐decadal coastal cliff retreat during the 21st century

USGS Publications Warehouse

Limber, Patrick; Barnard, Patrick; Vitousek, Sean; Erikson, Li

2018-01-01

Sea cliff retreat rates are expected to accelerate with rising sea levels during the 21st century. Here we develop an approach for a multi‐model ensemble that efficiently projects time‐averaged sea cliff retreat over multi‐decadal time scales and large (>50 km) spatial scales. The ensemble consists of five simple 1‐D models adapted from the literature that relate sea cliff retreat to wave impacts, sea level rise (SLR), historical cliff behavior, and cross‐shore profile geometry. Ensemble predictions are based on Monte Carlo simulations of each individual model, which account for the uncertainty of model parameters. The consensus of the individual models also weights uncertainty, such that uncertainty is greater when predictions from different models do not agree. A calibrated, but unvalidated, ensemble was applied to the 475 km‐long coastline of Southern California (USA), with 4 SLR scenarios of 0.5, 0.93, 1.5, and 2 m by 2100. Results suggest that future retreat rates could increase relative to mean historical rates by more than two‐fold for the higher SLR scenarios, causing an average total land loss of 19 – 41 m by 2100. However, model uncertainty ranges from +/‐ 5 – 15 m, reflecting the inherent difficulties of projecting cliff retreat over multiple decades. To enhance ensemble performance, future work could include weighting each model by its skill in matching observations in different morphological settings

The Limits of Coding with Joint Constraints on Detected and Undetected Error Rates

NASA Technical Reports Server (NTRS)

Dolinar, Sam; Andrews, Kenneth; Pollara, Fabrizio; Divsalar, Dariush

2008-01-01

We develop a remarkably tight upper bound on the performance of a parameterized family of bounded angle maximum-likelihood (BA-ML) incomplete decoders. The new bound for this class of incomplete decoders is calculated from the code's weight enumerator, and is an extension of Poltyrev-type bounds developed for complete ML decoders. This bound can also be applied to bound the average performance of random code ensembles in terms of an ensemble average weight enumerator. We also formulate conditions defining a parameterized family of optimal incomplete decoders, defined to minimize both the total codeword error probability and the undetected error probability for any fixed capability of the decoder to detect errors. We illustrate the gap between optimal and BA-ML incomplete decoding via simulation of a small code.

Enhanced Sampling in the Well-Tempered Ensemble

NASA Astrophysics Data System (ADS)

Bonomi, M.; Parrinello, M.

2010-05-01

We introduce the well-tempered ensemble (WTE) which is the biased ensemble sampled by well-tempered metadynamics when the energy is used as collective variable. WTE can be designed so as to have approximately the same average energy as the canonical ensemble but much larger fluctuations. These two properties lead to an extremely fast exploration of phase space. An even greater efficiency is obtained when WTE is combined with parallel tempering. Unbiased Boltzmann averages are computed on the fly by a recently developed reweighting method [M. Bonomi , J. Comput. Chem. 30, 1615 (2009)JCCHDD0192-865110.1002/jcc.21305]. We apply WTE and its parallel tempering variant to the 2d Ising model and to a Gō model of HIV protease, demonstrating in these two representative cases that convergence is accelerated by orders of magnitude.

Enhanced sampling in the well-tempered ensemble.

PubMed

Bonomi, M; Parrinello, M

2010-05-14

We introduce the well-tempered ensemble (WTE) which is the biased ensemble sampled by well-tempered metadynamics when the energy is used as collective variable. WTE can be designed so as to have approximately the same average energy as the canonical ensemble but much larger fluctuations. These two properties lead to an extremely fast exploration of phase space. An even greater efficiency is obtained when WTE is combined with parallel tempering. Unbiased Boltzmann averages are computed on the fly by a recently developed reweighting method [M. Bonomi, J. Comput. Chem. 30, 1615 (2009)]. We apply WTE and its parallel tempering variant to the 2d Ising model and to a Gō model of HIV protease, demonstrating in these two representative cases that convergence is accelerated by orders of magnitude.

Regional patterns of future runoff changes from Earth system models constrained by observation

NASA Astrophysics Data System (ADS)

Yang, Hui; Zhou, Feng; Piao, Shilong; Huang, Mengtian; Chen, Anping; Ciais, Philippe; Li, Yue; Lian, Xu; Peng, Shushi; Zeng, Zhenzhong

2017-06-01

In the recent Intergovernmental Panel on Climate Change assessment, multimodel ensembles (arithmetic model averaging, AMA) were constructed with equal weights given to Earth system models, without considering the performance of each model at reproducing current conditions. Here we use Bayesian model averaging (BMA) to construct a weighted model ensemble for runoff projections. Higher weights are given to models with better performance in estimating historical decadal mean runoff. Using the BMA method, we find that by the end of this century, the increase of global runoff (9.8 ± 1.5%) under Representative Concentration Pathway 8.5 is significantly lower than estimated from AMA (12.2 ± 1.3%). BMA presents a less severe runoff increase than AMA at northern high latitudes and a more severe decrease in Amazonia. Runoff decrease in Amazonia is stronger than the intermodel difference. The intermodel difference in runoff changes is mainly caused not only by precipitation differences among models, but also by evapotranspiration differences at the high northern latitudes.

Using Bayesian Model Averaging (BMA) to calibrate probabilistic surface temperature forecasts over Iran

NASA Astrophysics Data System (ADS)

Soltanzadeh, I.; Azadi, M.; Vakili, G. A.

2011-07-01

Using Bayesian Model Averaging (BMA), an attempt was made to obtain calibrated probabilistic numerical forecasts of 2-m temperature over Iran. The ensemble employs three limited area models (WRF, MM5 and HRM), with WRF used with five different configurations. Initial and boundary conditions for MM5 and WRF are obtained from the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) and for HRM the initial and boundary conditions come from analysis of Global Model Europe (GME) of the German Weather Service. The resulting ensemble of seven members was run for a period of 6 months (from December 2008 to May 2009) over Iran. The 48-h raw ensemble outputs were calibrated using BMA technique for 120 days using a 40 days training sample of forecasts and relative verification data. The calibrated probabilistic forecasts were assessed using rank histogram and attribute diagrams. Results showed that application of BMA improved the reliability of the raw ensemble. Using the weighted ensemble mean forecast as a deterministic forecast it was found that the deterministic-style BMA forecasts performed usually better than the best member's deterministic forecast.

Fitting a function to time-dependent ensemble averaged data.

PubMed

Fogelmark, Karl; Lomholt, Michael A; Irbäck, Anders; Ambjörnsson, Tobias

2018-05-03

Time-dependent ensemble averages, i.e., trajectory-based averages of some observable, are of importance in many fields of science. A crucial objective when interpreting such data is to fit these averages (for instance, squared displacements) with a function and extract parameters (such as diffusion constants). A commonly overlooked challenge in such function fitting procedures is that fluctuations around mean values, by construction, exhibit temporal correlations. We show that the only available general purpose function fitting methods, correlated chi-square method and the weighted least squares method (which neglects correlation), fail at either robust parameter estimation or accurate error estimation. We remedy this by deriving a new closed-form error estimation formula for weighted least square fitting. The new formula uses the full covariance matrix, i.e., rigorously includes temporal correlations, but is free of the robustness issues, inherent to the correlated chi-square method. We demonstrate its accuracy in four examples of importance in many fields: Brownian motion, damped harmonic oscillation, fractional Brownian motion and continuous time random walks. We also successfully apply our method, weighted least squares including correlation in error estimation (WLS-ICE), to particle tracking data. The WLS-ICE method is applicable to arbitrary fit functions, and we provide a publically available WLS-ICE software.

Robust electroencephalogram phase estimation with applications in brain-computer interface systems.

PubMed

Seraj, Esmaeil; Sameni, Reza

2017-03-01

In this study, a robust method is developed for frequency-specific electroencephalogram (EEG) phase extraction using the analytic representation of the EEG. Based on recent theoretical findings in this area, it is shown that some of the phase variations-previously associated to the brain response-are systematic side-effects of the methods used for EEG phase calculation, especially during low analytical amplitude segments of the EEG. With this insight, the proposed method generates randomized ensembles of the EEG phase using minor perturbations in the zero-pole loci of narrow-band filters, followed by phase estimation using the signal's analytical form and ensemble averaging over the randomized ensembles to obtain a robust EEG phase and frequency. This Monte Carlo estimation method is shown to be very robust to noise and minor changes of the filter parameters and reduces the effect of fake EEG phase jumps, which do not have a cerebral origin. As proof of concept, the proposed method is used for extracting EEG phase features for a brain computer interface (BCI) application. The results show significant improvement in classification rates using rather simple phase-related features and a standard K-nearest neighbors and random forest classifiers, over a standard BCI dataset. The average performance was improved between 4-7% (in absence of additive noise) and 8-12% (in presence of additive noise). The significance of these improvements was statistically confirmed by a paired sample t-test, with 0.01 and 0.03 p-values, respectively. The proposed method for EEG phase calculation is very generic and may be applied to other EEG phase-based studies.

Equilibrium energy spectrum of point vortex motion with remarks on ensemble choice and ergodicity

NASA Astrophysics Data System (ADS)

Esler, J. G.

2017-01-01

The dynamics and statistical mechanics of N chaotically evolving point vortices in the doubly periodic domain are revisited. The selection of the correct microcanonical ensemble for the system is first investigated. The numerical results of Weiss and McWilliams [Phys. Fluids A 3, 835 (1991), 10.1063/1.858014], who argued that the point vortex system with N =6 is nonergodic because of an apparent discrepancy between ensemble averages and dynamical time averages, are shown to be due to an incorrect ensemble definition. When the correct microcanonical ensemble is sampled, accounting for the vortex momentum constraint, time averages obtained from direct numerical simulation agree with ensemble averages within the sampling error of each calculation, i.e., there is no numerical evidence for nonergodicity. Further, in the N →∞ limit it is shown that the vortex momentum no longer constrains the long-time dynamics and therefore that the correct microcanonical ensemble for statistical mechanics is that associated with the entire constant energy hypersurface in phase space. Next, a recently developed technique is used to generate an explicit formula for the density of states function for the system, including for arbitrary distributions of vortex circulations. Exact formulas for the equilibrium energy spectrum, and for the probability density function of the energy in each Fourier mode, are then obtained. Results are compared with a series of direct numerical simulations with N =50 and excellent agreement is found, confirming the relevance of the results for interpretation of quantum and classical two-dimensional turbulence.

Large ensemble modeling of the last deglacial retreat of the West Antarctic Ice Sheet: comparison of simple and advanced statistical techniques

NASA Astrophysics Data System (ADS)

Pollard, David; Chang, Won; Haran, Murali; Applegate, Patrick; DeConto, Robert

2016-05-01

A 3-D hybrid ice-sheet model is applied to the last deglacial retreat of the West Antarctic Ice Sheet over the last ˜ 20 000 yr. A large ensemble of 625 model runs is used to calibrate the model to modern and geologic data, including reconstructed grounding lines, relative sea-level records, elevation-age data and uplift rates, with an aggregate score computed for each run that measures overall model-data misfit. Two types of statistical methods are used to analyze the large-ensemble results: simple averaging weighted by the aggregate score, and more advanced Bayesian techniques involving Gaussian process-based emulation and calibration, and Markov chain Monte Carlo. The analyses provide sea-level-rise envelopes with well-defined parametric uncertainty bounds, but the simple averaging method only provides robust results with full-factorial parameter sampling in the large ensemble. Results for best-fit parameter ranges and envelopes of equivalent sea-level rise with the simple averaging method agree well with the more advanced techniques. Best-fit parameter ranges confirm earlier values expected from prior model tuning, including large basal sliding coefficients on modern ocean beds.

A brief history of the introduction of generalized ensembles to Markov chain Monte Carlo simulations

NASA Astrophysics Data System (ADS)

Berg, Bernd A.

2017-03-01

The most efficient weights for Markov chain Monte Carlo calculations of physical observables are not necessarily those of the canonical ensemble. Generalized ensembles, which do not exist in nature but can be simulated on computers, lead often to a much faster convergence. In particular, they have been used for simulations of first order phase transitions and for simulations of complex systems in which conflicting constraints lead to a rugged free energy landscape. Starting off with the Metropolis algorithm and Hastings' extension, I present a minireview which focuses on the explosive use of generalized ensembles in the early 1990s. Illustrations are given, which range from spin models to peptides.

Identifying the optimal segmentors for mass classification in mammograms

NASA Astrophysics Data System (ADS)

Zhang, Yu; Tomuro, Noriko; Furst, Jacob; Raicu, Daniela S.

2015-03-01

In this paper, we present the results of our investigation on identifying the optimal segmentor(s) from an ensemble of weak segmentors, used in a Computer-Aided Diagnosis (CADx) system which classifies suspicious masses in mammograms as benign or malignant. This is an extension of our previous work, where we used various parameter settings of image enhancement techniques to each suspicious mass (region of interest (ROI)) to obtain several enhanced images, then applied segmentation to each image to obtain several contours of a given mass. Each segmentation in this ensemble is essentially a "weak segmentor" because no single segmentation can produce the optimal result for all images. Then after shape features are computed from the segmented contours, the final classification model was built using logistic regression. The work in this paper focuses on identifying the optimal segmentor(s) from an ensemble mix of weak segmentors. For our purpose, optimal segmentors are those in the ensemble mix which contribute the most to the overall classification rather than the ones that produced high precision segmentation. To measure the segmentors' contribution, we examined weights on the features in the derived logistic regression model and computed the average feature weight for each segmentor. The result showed that, while in general the segmentors with higher segmentation success rates had higher feature weights, some segmentors with lower segmentation rates had high classification feature weights as well.

Design of Probabilistic Random Forests with Applications to Anticancer Drug Sensitivity Prediction

PubMed Central

Rahman, Raziur; Haider, Saad; Ghosh, Souparno; Pal, Ranadip

2015-01-01

Random forests consisting of an ensemble of regression trees with equal weights are frequently used for design of predictive models. In this article, we consider an extension of the methodology by representing the regression trees in the form of probabilistic trees and analyzing the nature of heteroscedasticity. The probabilistic tree representation allows for analytical computation of confidence intervals (CIs), and the tree weight optimization is expected to provide stricter CIs with comparable performance in mean error. We approached the ensemble of probabilistic trees’ prediction from the perspectives of a mixture distribution and as a weighted sum of correlated random variables. We applied our methodology to the drug sensitivity prediction problem on synthetic and cancer cell line encyclopedia dataset and illustrated that tree weights can be selected to reduce the average length of the CI without increase in mean error. PMID:27081304

Probabilistic Near and Far-Future Climate Scenarios of Precipitation and Surface Temperature for the North American Monsoon Region Under a Weighted CMIP5-GCM Ensemble Approach.

NASA Astrophysics Data System (ADS)

Montero-Martinez, M. J.; Colorado, G.; Diaz-Gutierrez, D. E.; Salinas-Prieto, J. A.

2017-12-01

It is well known the North American Monsoon (NAM) region is already a very dry region which is under a lot of stress due to the lack of water resources on multiple locations of the area. However, it is very interesting that even under those conditions, the Mexican part of the NAM region is certainly the most productive in Mexico from the agricultural point of view. Thus, it is very important to have realistic climate scenarios for climate variables such as temperature, precipitation, relative humidity, radiation, etc. This study tries to tackle that problem by generating probabilistic climate scenarios using a weighted CMIP5-GCM ensemble approach based on the Xu et al. (2010) technique which is on itself an improved method from the better known Reliability Ensemble Averaging algorithm of Giorgi and Mearns (2002). In addition, it is compared the 20-plus GCMs individual performances and the weighted ensemble versus observed data (CRU TS2.1) by using different metrics and Taylor diagrams. This study focuses on probabilistic results reaching a certain threshold given the fact that those types of products could be of potential use for agricultural applications.

The Albedo of Kepler's Small Worlds

NASA Astrophysics Data System (ADS)

Jansen, Tiffany; Kipping, David

2018-01-01

The study of exoplanet phase curves has been established as a powerful tool for measuring the atmospheric properties of other worlds. To first order, phase curves have the same amplitude as occultations, yet far greater temporal baselines enabling substantial improvements in sensitivity. Even so, only a relatively small fraction of Kepler planets have detectable phase curves, leading to a population dominated by hot-Jupiters. One way to boost sensitivity further is to stack different planets of similar types together, giving rise to an average phase curve for a specific ensemble. In this work, we measure the average albedo, thermal redistribution efficiency, and greenhouse boosting factor from the average phase curves of 115 Neptunian and 50 Terran (solid) worlds. We construct ensemble phase curve models for both samples accounting for the reflection and thermal components and regress our models assuming a global albedo, redistribution factor and greenhouse factor in a Bayesian framework. We find modest evidence for a detected phase curve in the Neptunian sample, although the albedo and thermal properties are somewhat degenerate meaning we can only place an upper limit on the albedo of Ag < 0.23 and greenhouse factor of f < 1.40 to 95% confidence. As predicted theoretically, this confirms hot-Neptunes are darker than Neptune and Uranus. Additionally, we place a constraint on the albedo of solid, Terran worlds of Ag < 0.42 and f < 1.60 to 95% confidence, compatible with a dark Lunar-like surface.

Effects of correlations and fees in random multiplicative environments: Implications for portfolio management.

PubMed

Alper, Ofer; Somekh-Baruch, Anelia; Pirvandy, Oz; Schaps, Malka; Yaari, Gur

2017-08-01

Geometric Brownian motion (GBM) is frequently used to model price dynamics of financial assets, and a weighted average of multiple GBMs is commonly used to model a financial portfolio. Diversified portfolios can lead to an increased exponential growth compared to a single asset by effectively reducing the effective noise. The sum of GBM processes is no longer a log-normal process and has a complex statistical properties. The nonergodicity of the weighted average process results in constant degradation of the exponential growth from the ensemble average toward the time average. One way to stay closer to the ensemble average is to maintain a balanced portfolio: keep the relative weights of the different assets constant over time. To keep these proportions constant, whenever assets values change, it is necessary to rebalance their relative weights, exposing this strategy to fees (transaction costs). Two strategies that were suggested in the past for cases that involve fees are rebalance the portfolio periodically and rebalance it in a partial way. In this paper, we study these two strategies in the presence of correlations and fees. We show that using periodic and partial rebalance strategies, it is possible to maintain a steady exponential growth while minimizing the losses due to fees. We also demonstrate how these redistribution strategies perform in a phenomenal way on real-world market data, despite the fact that not all assumptions of the model hold in these real-world systems. Our results have important implications for stochastic dynamics in general and to portfolio management in particular, as we show that there is a superior alternative to the common buy-and-hold strategy, even in the presence of correlations and fees.

Effects of correlations and fees in random multiplicative environments: Implications for portfolio management

NASA Astrophysics Data System (ADS)

Alper, Ofer; Somekh-Baruch, Anelia; Pirvandy, Oz; Schaps, Malka; Yaari, Gur

2017-08-01

Geometric Brownian motion (GBM) is frequently used to model price dynamics of financial assets, and a weighted average of multiple GBMs is commonly used to model a financial portfolio. Diversified portfolios can lead to an increased exponential growth compared to a single asset by effectively reducing the effective noise. The sum of GBM processes is no longer a log-normal process and has a complex statistical properties. The nonergodicity of the weighted average process results in constant degradation of the exponential growth from the ensemble average toward the time average. One way to stay closer to the ensemble average is to maintain a balanced portfolio: keep the relative weights of the different assets constant over time. To keep these proportions constant, whenever assets values change, it is necessary to rebalance their relative weights, exposing this strategy to fees (transaction costs). Two strategies that were suggested in the past for cases that involve fees are rebalance the portfolio periodically and rebalance it in a partial way. In this paper, we study these two strategies in the presence of correlations and fees. We show that using periodic and partial rebalance strategies, it is possible to maintain a steady exponential growth while minimizing the losses due to fees. We also demonstrate how these redistribution strategies perform in a phenomenal way on real-world market data, despite the fact that not all assumptions of the model hold in these real-world systems. Our results have important implications for stochastic dynamics in general and to portfolio management in particular, as we show that there is a superior alternative to the common buy-and-hold strategy, even in the presence of correlations and fees.

Equipartition terms in transition path ensemble: Insights from molecular dynamics simulations of alanine dipeptide.

PubMed

Li, Wenjin

2018-02-28

Transition path ensemble consists of reactive trajectories and possesses all the information necessary for the understanding of the mechanism and dynamics of important condensed phase processes. However, quantitative description of the properties of the transition path ensemble is far from being established. Here, with numerical calculations on a model system, the equipartition terms defined in thermal equilibrium were for the first time estimated in the transition path ensemble. It was not surprising to observe that the energy was not equally distributed among all the coordinates. However, the energies distributed on a pair of conjugated coordinates remained equal. Higher energies were observed to be distributed on several coordinates, which are highly coupled to the reaction coordinate, while the rest were almost equally distributed. In addition, the ensemble-averaged energy on each coordinate as a function of time was also quantified. These quantitative analyses on energy distributions provided new insights into the transition path ensemble.

Multiphysics superensemble forecast applied to Mediterranean heavy precipitation situations

NASA Astrophysics Data System (ADS)

Vich, M.; Romero, R.

2010-11-01

The high-impact precipitation events that regularly affect the western Mediterranean coastal regions are still difficult to predict with the current prediction systems. Bearing this in mind, this paper focuses on the superensemble technique applied to the precipitation field. Encouraged by the skill shown by a previous multiphysics ensemble prediction system applied to western Mediterranean precipitation events, the superensemble is fed with this ensemble. The training phase of the superensemble contributes to the actual forecast with weights obtained by comparing the past performance of the ensemble members and the corresponding observed states. The non-hydrostatic MM5 mesoscale model is used to run the multiphysics ensemble. Simulations are performed with a 22.5 km resolution domain (Domain 1 in http://mm5forecasts.uib.es) nested in the ECMWF forecast fields. The period between September and December 2001 is used to train the superensemble and a collection of 19~MEDEX cyclones is used to test it. The verification procedure involves testing the superensemble performance and comparing it with that of the poor-man and bias-corrected ensemble mean and the multiphysic EPS control member. The results emphasize the need of a well-behaved training phase to obtain good results with the superensemble technique. A strategy to obtain this improved training phase is already outlined.

Impacts of weighting climate models for hydro-meteorological climate change studies

NASA Astrophysics Data System (ADS)

Chen, Jie; Brissette, François P.; Lucas-Picher, Philippe; Caya, Daniel

2017-06-01

Weighting climate models is controversial in climate change impact studies using an ensemble of climate simulations from different climate models. In climate science, there is a general consensus that all climate models should be considered as having equal performance or in other words that all projections are equiprobable. On the other hand, in the impacts and adaptation community, many believe that climate models should be weighted based on their ability to better represent various metrics over a reference period. The debate appears to be partly philosophical in nature as few studies have investigated the impact of using weights in projecting future climate changes. The present study focuses on the impact of assigning weights to climate models for hydrological climate change studies. Five methods are used to determine weights on an ensemble of 28 global climate models (GCMs) adapted from the Coupled Model Intercomparison Project Phase 5 (CMIP5) database. Using a hydrological model, streamflows are computed over a reference (1961-1990) and future (2061-2090) periods, with and without post-processing climate model outputs. The impacts of using different weighting schemes for GCM simulations are then analyzed in terms of ensemble mean and uncertainty. The results show that weighting GCMs has a limited impact on both projected future climate in term of precipitation and temperature changes and hydrology in terms of nine different streamflow criteria. These results apply to both raw and post-processed GCM model outputs, thus supporting the view that climate models should be considered equiprobable.

Annealed importance sampling with constant cooling rate

NASA Astrophysics Data System (ADS)

Giovannelli, Edoardo; Cardini, Gianni; Gellini, Cristina; Pietraperzia, Giangaetano; Chelli, Riccardo

2015-02-01

Annealed importance sampling is a simulation method devised by Neal [Stat. Comput. 11, 125 (2001)] to assign weights to configurations generated by simulated annealing trajectories. In particular, the equilibrium average of a generic physical quantity can be computed by a weighted average exploiting weights and estimates of this quantity associated to the final configurations of the annealed trajectories. Here, we review annealed importance sampling from the perspective of nonequilibrium path-ensemble averages [G. E. Crooks, Phys. Rev. E 61, 2361 (2000)]. The equivalence of Neal's and Crooks' treatments highlights the generality of the method, which goes beyond the mere thermal-based protocols. Furthermore, we show that a temperature schedule based on a constant cooling rate outperforms stepwise cooling schedules and that, for a given elapsed computer time, performances of annealed importance sampling are, in general, improved by increasing the number of intermediate temperatures.

Learning disordered topological phases by statistical recovery of symmetry

NASA Astrophysics Data System (ADS)

Yoshioka, Nobuyuki; Akagi, Yutaka; Katsura, Hosho

2018-05-01

We apply the artificial neural network in a supervised manner to map out the quantum phase diagram of disordered topological superconductors in class DIII. Given the disorder that keeps the discrete symmetries of the ensemble as a whole, translational symmetry which is broken in the quasiparticle distribution individually is recovered statistically by taking an ensemble average. By using this, we classify the phases by the artificial neural network that learned the quasiparticle distribution in the clean limit and show that the result is totally consistent with the calculation by the transfer matrix method or noncommutative geometry approach. If all three phases, namely the Z2, trivial, and thermal metal phases, appear in the clean limit, the machine can classify them with high confidence over the entire phase diagram. If only the former two phases are present, we find that the machine remains confused in a certain region, leading us to conclude the detection of the unknown phase which is eventually identified as the thermal metal phase.

Generalized ensemble method applied to study systems with strong first order transitions

DOE PAGES

Malolepsza, E.; Kim, J.; Keyes, T.

2015-09-28

At strong first-order phase transitions, the entropy versus energy or, at constant pressure, enthalpy, exhibits convex behavior, and the statistical temperature curve correspondingly exhibits an S-loop or back-bending. In the canonical and isothermal-isobaric ensembles, with temperature as the control variable, the probability density functions become bimodal with peaks localized outside of the S-loop region. Inside, states are unstable, and as a result simulation of equilibrium phase coexistence becomes impossible. To overcome this problem, a method was proposed by Kim, Keyes and Straub, where optimally designed generalized ensemble sampling was combined with replica exchange, and denoted generalized replica exchange method (gREM).more » This new technique uses parametrized effective sampling weights that lead to a unimodal energy distribution, transforming unstable states into stable ones. In the present study, the gREM, originally developed as a Monte Carlo algorithm, was implemented to work with molecular dynamics in an isobaric ensemble and coded into LAMMPS, a highly optimized open source molecular simulation package. Lastly, the method is illustrated in a study of the very strong solid/liquid transition in water.« less

Generalized ensemble method applied to study systems with strong first order transitions

NASA Astrophysics Data System (ADS)

Małolepsza, E.; Kim, J.; Keyes, T.

2015-09-01

At strong first-order phase transitions, the entropy versus energy or, at constant pressure, enthalpy, exhibits convex behavior, and the statistical temperature curve correspondingly exhibits an S-loop or back-bending. In the canonical and isothermal-isobaric ensembles, with temperature as the control variable, the probability density functions become bimodal with peaks localized outside of the S-loop region. Inside, states are unstable, and as a result simulation of equilibrium phase coexistence becomes impossible. To overcome this problem, a method was proposed by Kim, Keyes and Straub [1], where optimally designed generalized ensemble sampling was combined with replica exchange, and denoted generalized replica exchange method (gREM). This new technique uses parametrized effective sampling weights that lead to a unimodal energy distribution, transforming unstable states into stable ones. In the present study, the gREM, originally developed as a Monte Carlo algorithm, was implemented to work with molecular dynamics in an isobaric ensemble and coded into LAMMPS, a highly optimized open source molecular simulation package. The method is illustrated in a study of the very strong solid/liquid transition in water.

Climate Model Ensemble Methodology: Rationale and Challenges

NASA Astrophysics Data System (ADS)

Vezer, M. A.; Myrvold, W.

2012-12-01

A tractable model of the Earth's atmosphere, or, indeed, any large, complex system, is inevitably unrealistic in a variety of ways. This will have an effect on the model's output. Nonetheless, we want to be able to rely on certain features of the model's output in studies aiming to detect, attribute, and project climate change. For this, we need assurance that these features reflect the target system, and are not artifacts of the unrealistic assumptions that go into the model. One technique for overcoming these limitations is to study ensembles of models which employ different simplifying assumptions and different methods of modelling. One then either takes as reliable certain outputs on which models in the ensemble agree, or takes the average of these outputs as the best estimate. Since the Intergovernmental Panel on Climate Change's Fourth Assessment Report (IPCC AR4) modellers have aimed to improve ensemble analysis by developing techniques to account for dependencies among models, and to ascribe unequal weights to models according to their performance. The goal of this paper is to present as clearly and cogently as possible the rationale for climate model ensemble methodology, the motivation of modellers to account for model dependencies, and their efforts to ascribe unequal weights to models. The method of our analysis is as follows. We will consider a simpler, well-understood case of taking the mean of a number of measurements of some quantity. Contrary to what is sometimes said, it is not a requirement of this practice that the errors of the component measurements be independent; one must, however, compensate for any lack of independence. We will also extend the usual accounts to include cases of unknown systematic error. We draw parallels between this simpler illustration and the more complex example of climate model ensembles, detailing how ensembles can provide more useful information than any of their constituent models. This account emphasizes the epistemic importance of considering degrees of model dependence, and the practice of ascribing unequal weights to models of unequal skill.

Ensemble hydro-meteorological forecasting for early warning of floods and scheduling of hydropower production

NASA Astrophysics Data System (ADS)

Solvang Johansen, Stian; Steinsland, Ingelin; Engeland, Kolbjørn

2016-04-01

Running hydrological models with precipitation and temperature ensemble forcing to generate ensembles of streamflow is a commonly used method in operational hydrology. Evaluations of streamflow ensembles have however revealed that the ensembles are biased with respect to both mean and spread. Thus postprocessing of the ensembles is needed in order to improve the forecast skill. The aims of this study is (i) to to evaluate how postprocessing of streamflow ensembles works for Norwegian catchments within different hydrological regimes and to (ii) demonstrate how post processed streamflow ensembles are used operationally by a hydropower producer. These aims were achieved by postprocessing forecasted daily discharge for 10 lead-times for 20 catchments in Norway by using EPS forcing from ECMWF applied the semi-distributed HBV-model dividing each catchment into 10 elevation zones. Statkraft Energi uses forecasts from these catchments for scheduling hydropower production. The catchments represent different hydrological regimes. Some catchments have stable winter condition with winter low flow and a major flood event during spring or early summer caused by snow melting. Others has a more mixed snow-rain regime, often with a secondary flood season during autumn, and in the coastal areas, the stream flow is dominated by rain, and the main flood season is autumn and winter. For post processing, a Bayesian model averaging model (BMA) close to (Kleiber et al 2011) is used. The model creates a predictive PDF that is a weighted average of PDFs centered on the individual bias corrected forecasts. The weights are here equal since all ensemble members come from the same model, and thus have the same probability. For modeling streamflow, the gamma distribution is chosen as a predictive PDF. The bias correction parameters and the PDF parameters are estimated using a 30-day sliding window training period. Preliminary results show that the improvement varies between catchments depending on where they are situated and the hydrological regime. There is an improvement in CRPS for all catchments compared to raw EPS ensembles. The improvement is up to lead-time 5-7. The postprocessing also improves the MAE for the median of the predictive PDF compared to the median of the raw EPS. But less compared to CRPS, often up to lead-time 2-3. The streamflow ensembles are to some extent used operationally in Statkraft Energi (Hydro Power company, Norway), with respect to early warning, risk assessment and decision-making. Presently all forecast used operationally for short-term scheduling are deterministic, but ensembles are used visually for expert assessment of risk in difficult situations where e.g. there is a chance of overflow in a reservoir. However, there are plans to incorporate ensembles in the daily scheduling of hydropower production.

Enhancing Flood Prediction Reliability Using Bayesian Model Averaging

NASA Astrophysics Data System (ADS)

Liu, Z.; Merwade, V.

2017-12-01

Uncertainty analysis is an indispensable part of modeling the hydrology and hydrodynamics of non-idealized environmental systems. Compared to reliance on prediction from one model simulation, using on ensemble of predictions that consider uncertainty from different sources is more reliable. In this study, Bayesian model averaging (BMA) is applied to Black River watershed in Arkansas and Missouri by combining multi-model simulations to get reliable deterministic water stage and probabilistic inundation extent predictions. The simulation ensemble is generated from 81 LISFLOOD-FP subgrid model configurations that include uncertainty from channel shape, channel width, channel roughness and discharge. Model simulation outputs are trained with observed water stage data during one flood event, and BMA prediction ability is validated for another flood event. Results from this study indicate that BMA does not always outperform all members in the ensemble, but it provides relatively robust deterministic flood stage predictions across the basin. Station based BMA (BMA_S) water stage prediction has better performance than global based BMA (BMA_G) prediction which is superior to the ensemble mean prediction. Additionally, high-frequency flood inundation extent (probability greater than 60%) in BMA_G probabilistic map is more accurate than the probabilistic flood inundation extent based on equal weights.

Improving ECG Classification Accuracy Using an Ensemble of Neural Network Modules

PubMed Central

Javadi, Mehrdad; Ebrahimpour, Reza; Sajedin, Atena; Faridi, Soheil; Zakernejad, Shokoufeh

2011-01-01

This paper illustrates the use of a combined neural network model based on Stacked Generalization method for classification of electrocardiogram (ECG) beats. In conventional Stacked Generalization method, the combiner learns to map the base classifiers' outputs to the target data. We claim adding the input pattern to the base classifiers' outputs helps the combiner to obtain knowledge about the input space and as the result, performs better on the same task. Experimental results support our claim that the additional knowledge according to the input space, improves the performance of the proposed method which is called Modified Stacked Generalization. In particular, for classification of 14966 ECG beats that were not previously seen during training phase, the Modified Stacked Generalization method reduced the error rate for 12.41% in comparison with the best of ten popular classifier fusion methods including Max, Min, Average, Product, Majority Voting, Borda Count, Decision Templates, Weighted Averaging based on Particle Swarm Optimization and Stacked Generalization. PMID:22046232

On averaging aspect ratios and distortion parameters over ice crystal population ensembles for estimating effective scattering asymmetry parameters

PubMed Central

van Diedenhoven, Bastiaan; Ackerman, Andrew S.; Fridlind, Ann M.; Cairns, Brian

2017-01-01

The use of ensemble-average values of aspect ratio and distortion parameter of hexagonal ice prisms for the estimation of ensemble-average scattering asymmetry parameters is evaluated. Using crystal aspect ratios greater than unity generally leads to ensemble-average values of aspect ratio that are inconsistent with the ensemble-average asymmetry parameters. When a definition of aspect ratio is used that limits the aspect ratio to below unity (α≤1) for both hexagonal plates and columns, the effective asymmetry parameters calculated using ensemble-average aspect ratios are generally consistent with ensemble-average asymmetry parameters, especially if aspect ratios are geometrically averaged. Ensemble-average distortion parameters generally also yield effective asymmetry parameters that are largely consistent with ensemble-average asymmetry parameters. In the case of mixtures of plates and columns, it is recommended to geometrically average the α≤1 aspect ratios and to subsequently calculate the effective asymmetry parameter using a column or plate geometry when the contribution by columns to a given mixture’s total projected area is greater or lower than 50%, respectively. In addition, we show that ensemble-average aspect ratios, distortion parameters and asymmetry parameters can generally be retrieved accurately from simulated multi-directional polarization measurements based on mixtures of varying columns and plates. However, such retrievals tend to be somewhat biased toward yielding column-like aspect ratios. Furthermore, generally large retrieval errors can occur for mixtures with approximately equal contributions of columns and plates and for ensembles with strong contributions of thin plates. PMID:28983127

Ergodicity of financial indices

NASA Astrophysics Data System (ADS)

Kolesnikov, A. V.; Rühl, T.

2010-05-01

We introduce the concept of the ensemble averaging for financial markets. We address the question of equality of ensemble and time averaging in their sequence and investigate if these averagings are equivalent for large amount of equity indices and branches. We start with the model of Gaussian-distributed returns, equal-weighted stocks in each index and absence of correlations within a single day and show that even this oversimplified model captures already the run of the corresponding index reasonably well due to its self-averaging properties. We introduce the concept of the instant cross-sectional volatility and discuss its relation to the ordinary time-resolved counterpart. The role of the cross-sectional volatility for the description of the corresponding index as well as the role of correlations between the single stocks and the role of non-Gaussianity of stock distributions is briefly discussed. Our model reveals quickly and efficiently some anomalies or bubbles in a particular financial market and gives an estimate of how large these effects can be and how quickly they disappear.

Synchronization Experiments With A Global Coupled Model of Intermediate Complexity

NASA Astrophysics Data System (ADS)

Selten, Frank; Hiemstra, Paul; Shen, Mao-Lin

2013-04-01

In the super modeling approach an ensemble of imperfect models are connected through nudging terms that nudge the solution of each model to the solution of all other models in the ensemble. The goal is to obtain a synchronized state through a proper choice of connection strengths that closely tracks the trajectory of the true system. For the super modeling approach to be successful, the connections should be dense and strong enough for synchronization to occur. In this study we analyze the behavior of an ensemble of connected global atmosphere-ocean models of intermediate complexity. All atmosphere models are connected to the same ocean model through the surface fluxes of heat, water and momentum, the ocean is integrated using weighted averaged surface fluxes. In particular we analyze the degree of synchronization between the atmosphere models and the characteristics of the ensemble mean solution. The results are interpreted using a low order atmosphere-ocean toy model.

Bayesian Ensemble Trees (BET) for Clustering and Prediction in Heterogeneous Data

PubMed Central

Duan, Leo L.; Clancy, John P.; Szczesniak, Rhonda D.

2016-01-01

We propose a novel “tree-averaging” model that utilizes the ensemble of classification and regression trees (CART). Each constituent tree is estimated with a subset of similar data. We treat this grouping of subsets as Bayesian Ensemble Trees (BET) and model them as a Dirichlet process. We show that BET determines the optimal number of trees by adapting to the data heterogeneity. Compared with the other ensemble methods, BET requires much fewer trees and shows equivalent prediction accuracy using weighted averaging. Moreover, each tree in BET provides variable selection criterion and interpretation for each subset. We developed an efficient estimating procedure with improved estimation strategies in both CART and mixture models. We demonstrate these advantages of BET with simulations and illustrate the approach with a real-world data example involving regression of lung function measurements obtained from patients with cystic fibrosis. Supplemental materials are available online. PMID:27524872

An ensemble rank learning approach for gene prioritization.

PubMed

Lee, Po-Feng; Soo, Von-Wun

2013-01-01

Several different computational approaches have been developed to solve the gene prioritization problem. We intend to use the ensemble boosting learning techniques to combine variant computational approaches for gene prioritization in order to improve the overall performance. In particular we add a heuristic weighting function to the Rankboost algorithm according to: 1) the absolute ranks generated by the adopted methods for a certain gene, and 2) the ranking relationship between all gene-pairs from each prioritization result. We select 13 known prostate cancer genes in OMIM database as training set and protein coding gene data in HGNC database as test set. We adopt the leave-one-out strategy for the ensemble rank boosting learning. The experimental results show that our ensemble learning approach outperforms the four gene-prioritization methods in ToppGene suite in the ranking results of the 13 known genes in terms of mean average precision, ROC and AUC measures.

--No Title--

Science.gov Websites

2008073000 2008072900 2008072800 Background information bias reduction = ( | domain-averaged ensemble mean bias | - | domain-averaged bias-corrected ensemble mean bias | / | domain-averaged bias-corrected ensemble mean bias | NAEFS Products | NAEFS | EMC Ensemble Products EMC | NCEP | National Weather Service

An information-theoretical perspective on weighted ensemble forecasts

NASA Astrophysics Data System (ADS)

Weijs, Steven V.; van de Giesen, Nick

2013-08-01

This paper presents an information-theoretical method for weighting ensemble forecasts with new information. Weighted ensemble forecasts can be used to adjust the distribution that an existing ensemble of time series represents, without modifying the values in the ensemble itself. The weighting can, for example, add new seasonal forecast information in an existing ensemble of historically measured time series that represents climatic uncertainty. A recent article in this journal compared several methods to determine the weights for the ensemble members and introduced the pdf-ratio method. In this article, a new method, the minimum relative entropy update (MRE-update), is presented. Based on the principle of minimum discrimination information, an extension of the principle of maximum entropy (POME), the method ensures that no more information is added to the ensemble than is present in the forecast. This is achieved by minimizing relative entropy, with the forecast information imposed as constraints. From this same perspective, an information-theoretical view on the various weighting methods is presented. The MRE-update is compared with the existing methods and the parallels with the pdf-ratio method are analysed. The paper provides a new, information-theoretical justification for one version of the pdf-ratio method that turns out to be equivalent to the MRE-update. All other methods result in sets of ensemble weights that, seen from the information-theoretical perspective, add either too little or too much (i.e. fictitious) information to the ensemble.

Finite-size anomalies of the Drude weight: Role of symmetries and ensembles

NASA Astrophysics Data System (ADS)

Sánchez, R. J.; Varma, V. K.

2017-12-01

We revisit the numerical problem of computing the high temperature spin stiffness, or Drude weight, D of the spin-1 /2 X X Z chain using exact diagonalization to systematically analyze its dependence on system symmetries and ensemble. Within the canonical ensemble and for states with zero total magnetization, we find D vanishes exactly due to spin-inversion symmetry for all but the anisotropies Δ˜M N=cos(π M /N ) with N ,M ∈Z+ coprimes and N >M , provided system sizes L ≥2 N , for which states with different spin-inversion signature become degenerate due to the underlying s l2 loop algebra symmetry. All these loop-algebra degenerate states carry finite currents which we conjecture [based on data from the system sizes and anisotropies Δ˜M N (with N

Using Bayes Model Averaging for Wind Power Forecasts

NASA Astrophysics Data System (ADS)

Preede Revheim, Pål; Beyer, Hans Georg

2014-05-01

For operational purposes predictions of the forecasts of the lumped output of groups of wind farms spread over larger geographic areas will often be of interest. A naive approach is to make forecasts for each individual site and sum them up to get the group forecast. It is however well documented that a better choice is to use a model that also takes advantage of spatial smoothing effects. It might however be the case that some sites tends to more accurately reflect the total output of the region, either in general or for certain wind directions. It will then be of interest giving these a greater influence over the group forecast. Bayesian model averaging (BMA) is a statistical post-processing method for producing probabilistic forecasts from ensembles. Raftery et al. [1] show how BMA can be used for statistical post processing of forecast ensembles, producing PDFs of future weather quantities. The BMA predictive PDF of a future weather quantity is a weighted average of the ensemble members' PDFs, where the weights can be interpreted as posterior probabilities and reflect the ensemble members' contribution to overall forecasting skill over a training period. In Revheim and Beyer [2] the BMA procedure used in Sloughter, Gneiting and Raftery [3] were found to produce fairly accurate PDFs for the future mean wind speed of a group of sites from the single sites wind speeds. However, when the procedure was attempted applied to wind power it resulted in either problems with the estimation of the parameters (mainly caused by longer consecutive periods of no power production) or severe underestimation (mainly caused by problems with reflecting the power curve). In this paper the problems that arose when applying BMA to wind power forecasting is met through two strategies. First, the BMA procedure is run with a combination of single site wind speeds and single site wind power production as input. This solves the problem with longer consecutive periods where the input data does not contain information, but it has the disadvantage of nearly doubling the number of model parameters to be estimated. Second, the BMA procedure is run with group mean wind power as the response variable instead of group mean wind speed. This also solves the problem with longer consecutive periods without information in the input data, but it leaves the power curve to also be estimated from the data. [1] Raftery, A. E., et al. (2005). Using Bayesian Model Averaging to Calibrate Forecast Ensembles. Monthly Weather Review, 133, 1155-1174. [2]Revheim, P. P. and H. G. Beyer (2013). Using Bayesian Model Averaging for wind farm group forecasts. EWEA Wind Power Forecasting Technology Workshop,Rotterdam, 4-5 December 2013. [3]Sloughter, J. M., T. Gneiting and A. E. Raftery (2010). Probabilistic Wind Speed Forecasting Using Ensembles and Bayesian Model Averaging. Journal of the American Statistical Association, Vol. 105, No. 489, 25-35

Single Aerosol Particle Studies Using Optical Trapping Raman And Cavity Ringdown Spectroscopy

NASA Astrophysics Data System (ADS)

Gong, Z.; Wang, C.; Pan, Y. L.; Videen, G.

2017-12-01

Due to the physical and chemical complexity of aerosol particles and the interdisciplinary nature of aerosol science that involves physics, chemistry, and biology, our knowledge of aerosol particles is rather incomplete; our current understanding of aerosol particles is limited by averaged (over size, composition, shape, and orientation) and/or ensemble (over time, size, and multi-particles) measurements. Physically, single aerosol particles are the fundamental units of any large aerosol ensembles. Chemically, single aerosol particles carry individual chemical components (properties and constituents) in particle ensemble processes. Therefore, the study of single aerosol particles can bridge the gap between aerosol ensembles and bulk/surface properties and provide a hierarchical progression from a simple benchmark single-component system to a mixed-phase multicomponent system. A single aerosol particle can be an effective reactor to study heterogeneous surface chemistry in multiple phases. Latest technological advances provide exciting new opportunities to study single aerosol particles and to further develop single aerosol particle instrumentation. We present updates on our recent studies of single aerosol particles optically trapped in air using the optical-trapping Raman and cavity ringdown spectroscopy.

Interactive vs. Non-Interactive Ensembles for Weather Prediction and Climate Projection

NASA Astrophysics Data System (ADS)

Duane, Gregory

2013-04-01

If the members of an ensemble of different models are allowed to interact with one another in run time, predictive skill can be improved as compared to that of any individual model or any average of indvidual model outputs. Inter-model connections in such an interactive ensemble can be trained, using historical data, so that the resulting ``supermodel" synchronizes with reality when used in weather-prediction mode, where the individual models perform data assimilation from each other (with trainable inter-model "observation error") as well as from real observations. In climate-projection mode, parameters of the individual models are changed, as might occur from an increase in GHG levels, and one obtains relevant statistical properties of the new supermodel attractor. In simple cases, it has been shown that training of the inter-model connections with the old parameter values gives a supermodel that is still predictive when the parameter values are changed. Here we inquire as to the circumstances under which supermodel performance can be expected to exceed that of the customary weighted average of model outputs. We consider a supermodel formed from quasigeostrophic channel models with different forcing coefficients, and introduce an effective training scheme for the inter-model connections. We show that the blocked-zonal index cycle is reproduced better by the supermodel than by any non-interactive ensemble in the extreme case where the forcing coefficients of the different models are very large or very small. With realistic differences in forcing coefficients, as would be representative of actual differences among IPCC-class models, the usual linearity assumption is justified and a weighted average of model outputs is adequate. It is therefore hypothesized that supermodeling is likely to be useful in situations where there are qualitative model differences, as arising from sub-gridscale parameterizations, that affect overall model behavior. Otherwise the usual ex post facto averaging will probably suffice. Previous results from an ENSO-prediction supermodel [Kirtman et al.] are re-examined in light of the hypothesis about the importance of qualitative inter-model differences.

Significance of model credibility in estimating climate projection distributions for regional hydroclimatological risk assessments

USGS Publications Warehouse

Brekke, L.D.; Dettinger, M.D.; Maurer, E.P.; Anderson, M.

2008-01-01

Ensembles of historical climate simulations and climate projections from the World Climate Research Programme's (WCRP's) Coupled Model Intercomparison Project phase 3 (CMIP3) multi-model dataset were investigated to determine how model credibility affects apparent relative scenario likelihoods in regional risk assessments. Methods were developed and applied in a Northern California case study. An ensemble of 59 twentieth century climate simulations from 17 WCRP CMIP3 models was analyzed to evaluate relative model credibility associated with a 75-member projection ensemble from the same 17 models. Credibility was assessed based on how models realistically reproduced selected statistics of historical climate relevant to California climatology. Metrics of this credibility were used to derive relative model weights leading to weight-threshold culling of models contributing to the projection ensemble. Density functions were then estimated for two projected quantities (temperature and precipitation), with and without considering credibility-based ensemble reductions. An analysis for Northern California showed that, while some models seem more capable at recreating limited aspects twentieth century climate, the overall tendency is for comparable model performance when several credibility measures are combined. Use of these metrics to decide which models to include in density function development led to local adjustments to function shapes, but led to limited affect on breadth and central tendency, which were found to be more influenced by 'completeness' of the original ensemble in terms of models and emissions pathways. ?? 2007 Springer Science+Business Media B.V.

Optimal weighted averaging of event related activity from acquisitions with artifacts.

PubMed

Vollero, Luca; Petrichella, Sara; Innello, Giulio

2016-08-01

In several biomedical applications that require the signal processing of biological data, the starting procedure for noise reduction is the ensemble averaging of multiple repeated acquisitions (trials). This method is based on the assumption that each trial is composed of two additive components: (i) a time-locked activity related to some sensitive/stimulation phenomenon (ERA, Event Related Activity in the following) and (ii) a sum of several other non time-locked background activities. The averaging aims at estimating the ERA activity under very low Signal to Noise and Interference Ratio (SNIR). Although averaging is a well established tool, its performance can be improved in the presence of high-power disturbances (artifacts) by a trials classification and removal stage. In this paper we propose, model and evaluate a new approach that avoids trials removal, managing trials classified as artifact-free and artifact-prone with two different weights. Based on the model, a weights tuning is possible and through modeling and simulations we show that, when optimally configured, the proposed solution outperforms classical approaches.

--No Title--

Science.gov Websites

2008112500 2008112400 Background information bias reduction = ( | domain-averaged ensemble mean bias | - | domain-averaged bias-corrected ensemble mean bias | / | domain-averaged bias-corrected ensemble mean bias

Thermal density functional theory, ensemble density functional theory, and potential functional theory for warm dense matter

NASA Astrophysics Data System (ADS)

Pribram-Jones, Aurora

Warm dense matter (WDM) is a high energy phase between solids and plasmas, with characteristics of both. It is present in the centers of giant planets, within the earth's core, and on the path to ignition of inertial confinement fusion. The high temperatures and pressures of warm dense matter lead to complications in its simulation, as both classical and quantum effects must be included. One of the most successful simulation methods is density functional theory-molecular dynamics (DFT-MD). Despite great success in a diverse array of applications, DFT-MD remains computationally expensive and it neglects the explicit temperature dependence of electron-electron interactions known to exist within exact DFT. Finite-temperature density functional theory (FT DFT) is an extension of the wildly successful ground-state DFT formalism via thermal ensembles, broadening its quantum mechanical treatment of electrons to include systems at non-zero temperatures. Exact mathematical conditions have been used to predict the behavior of approximations in limiting conditions and to connect FT DFT to the ground-state theory. An introduction to FT DFT is given within the context of ensemble DFT and the larger field of DFT is discussed for context. Ensemble DFT is used to describe ensembles of ground-state and excited systems. Exact conditions in ensemble DFT and the performance of approximations depend on ensemble weights. Using an inversion method, exact Kohn-Sham ensemble potentials are found and compared to approximations. The symmetry eigenstate Hartree-exchange approximation is in good agreement with exact calculations because of its inclusion of an ensemble derivative discontinuity. Since ensemble weights in FT DFT are temperature-dependent Fermi weights, this insight may help develop approximations well-suited to both ground-state and FT DFT. A novel, highly efficient approach to free energy calculations, finite-temperature potential functional theory, is derived, which has the potential to transform the simulation of warm dense matter. As a semiclassical method, it connects the normally disparate regimes of cold condensed matter physics and hot plasma physics. This orbital-free approach captures the smooth classical density envelope and quantum density oscillations that are both crucial to accurate modeling of materials where temperature and pressure effects are influential.

Selecting a Classification Ensemble and Detecting Process Drift in an Evolving Data Stream

DOE Office of Scientific and Technical Information (OSTI.GOV)

Heredia-Langner, Alejandro; Rodriguez, Luke R.; Lin, Andy

2015-09-30

We characterize the commercial behavior of a group of companies in a common line of business using a small ensemble of classifiers on a stream of records containing commercial activity information. This approach is able to effectively find a subset of classifiers that can be used to predict company labels with reasonable accuracy. Performance of the ensemble, its error rate under stable conditions, can be characterized using an exponentially weighted moving average (EWMA) statistic. The behavior of the EWMA statistic can be used to monitor a record stream from the commercial network and determine when significant changes have occurred. Resultsmore » indicate that larger classification ensembles may not necessarily be optimal, pointing to the need to search the combinatorial classifier space in a systematic way. Results also show that current and past performance of an ensemble can be used to detect when statistically significant changes in the activity of the network have occurred. The dataset used in this work contains tens of thousands of high level commercial activity records with continuous and categorical variables and hundreds of labels, making classification challenging.« less

Bayesian model averaging using particle filtering and Gaussian mixture modeling: Theory, concepts, and simulation experiments

NASA Astrophysics Data System (ADS)

Rings, Joerg; Vrugt, Jasper A.; Schoups, Gerrit; Huisman, Johan A.; Vereecken, Harry

2012-05-01

Bayesian model averaging (BMA) is a standard method for combining predictive distributions from different models. In recent years, this method has enjoyed widespread application and use in many fields of study to improve the spread-skill relationship of forecast ensembles. The BMA predictive probability density function (pdf) of any quantity of interest is a weighted average of pdfs centered around the individual (possibly bias-corrected) forecasts, where the weights are equal to posterior probabilities of the models generating the forecasts, and reflect the individual models skill over a training (calibration) period. The original BMA approach presented by Raftery et al. (2005) assumes that the conditional pdf of each individual model is adequately described with a rather standard Gaussian or Gamma statistical distribution, possibly with a heteroscedastic variance. Here we analyze the advantages of using BMA with a flexible representation of the conditional pdf. A joint particle filtering and Gaussian mixture modeling framework is presented to derive analytically, as closely and consistently as possible, the evolving forecast density (conditional pdf) of each constituent ensemble member. The median forecasts and evolving conditional pdfs of the constituent models are subsequently combined using BMA to derive one overall predictive distribution. This paper introduces the theory and concepts of this new ensemble postprocessing method, and demonstrates its usefulness and applicability by numerical simulation of the rainfall-runoff transformation using discharge data from three different catchments in the contiguous United States. The revised BMA method receives significantly lower-prediction errors than the original default BMA method (due to filtering) with predictive uncertainty intervals that are substantially smaller but still statistically coherent (due to the use of a time-variant conditional pdf).

Shear-stress fluctuations and relaxation in polymer glasses

NASA Astrophysics Data System (ADS)

Kriuchevskyi, I.; Wittmer, J. P.; Meyer, H.; Benzerara, O.; Baschnagel, J.

2018-01-01

We investigate by means of molecular dynamics simulation a coarse-grained polymer glass model focusing on (quasistatic and dynamical) shear-stress fluctuations as a function of temperature T and sampling time Δ t . The linear response is characterized using (ensemble-averaged) expectation values of the contributions (time averaged for each shear plane) to the stress-fluctuation relation μsf for the shear modulus and the shear-stress relaxation modulus G (t ) . Using 100 independent configurations, we pay attention to the respective standard deviations. While the ensemble-averaged modulus μsf(T ) decreases continuously with increasing T for all Δ t sampled, its standard deviation δ μsf(T ) is nonmonotonic with a striking peak at the glass transition. The question of whether the shear modulus is continuous or has a jump singularity at the glass transition is thus ill posed. Confirming the effective time-translational invariance of our systems, the Δ t dependence of μsf and related quantities can be understood using a weighted integral over G (t ) .

Individual differences in ensemble perception reveal multiple, independent levels of ensemble representation.

PubMed

Haberman, Jason; Brady, Timothy F; Alvarez, George A

2015-04-01

Ensemble perception, including the ability to "see the average" from a group of items, operates in numerous feature domains (size, orientation, speed, facial expression, etc.). Although the ubiquity of ensemble representations is well established, the large-scale cognitive architecture of this process remains poorly defined. We address this using an individual differences approach. In a series of experiments, observers saw groups of objects and reported either a single item from the group or the average of the entire group. High-level ensemble representations (e.g., average facial expression) showed complete independence from low-level ensemble representations (e.g., average orientation). In contrast, low-level ensemble representations (e.g., orientation and color) were correlated with each other, but not with high-level ensemble representations (e.g., facial expression and person identity). These results suggest that there is not a single domain-general ensemble mechanism, and that the relationship among various ensemble representations depends on how proximal they are in representational space. (c) 2015 APA, all rights reserved).

Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping.

PubMed

Shafizadeh-Moghadam, Hossein; Valavi, Roozbeh; Shahabi, Himan; Chapi, Kamran; Shirzadi, Ataollah

2018-07-01

In this research, eight individual machine learning and statistical models are implemented and compared, and based on their results, seven ensemble models for flood susceptibility assessment are introduced. The individual models included artificial neural networks, classification and regression trees, flexible discriminant analysis, generalized linear model, generalized additive model, boosted regression trees, multivariate adaptive regression splines, and maximum entropy, and the ensemble models were Ensemble Model committee averaging (EMca), Ensemble Model confidence interval Inferior (EMciInf), Ensemble Model confidence interval Superior (EMciSup), Ensemble Model to estimate the coefficient of variation (EMcv), Ensemble Model to estimate the mean (EMmean), Ensemble Model to estimate the median (EMmedian), and Ensemble Model based on weighted mean (EMwmean). The data set covered 201 flood events in the Haraz watershed (Mazandaran province in Iran) and 10,000 randomly selected non-occurrence points. Among the individual models, the Area Under the Receiver Operating Characteristic (AUROC), which showed the highest value, belonged to boosted regression trees (0.975) and the lowest value was recorded for generalized linear model (0.642). On the other hand, the proposed EMmedian resulted in the highest accuracy (0.976) among all models. In spite of the outstanding performance of some models, nevertheless, variability among the prediction of individual models was considerable. Therefore, to reduce uncertainty, creating more generalizable, more stable, and less sensitive models, ensemble forecasting approaches and in particular the EMmedian is recommended for flood susceptibility assessment. Copyright © 2018 Elsevier Ltd. All rights reserved.

Girsanov reweighting for path ensembles and Markov state models

NASA Astrophysics Data System (ADS)

Donati, L.; Hartmann, C.; Keller, B. G.

2017-06-01

The sensitivity of molecular dynamics on changes in the potential energy function plays an important role in understanding the dynamics and function of complex molecules. We present a method to obtain path ensemble averages of a perturbed dynamics from a set of paths generated by a reference dynamics. It is based on the concept of path probability measure and the Girsanov theorem, a result from stochastic analysis to estimate a change of measure of a path ensemble. Since Markov state models (MSMs) of the molecular dynamics can be formulated as a combined phase-space and path ensemble average, the method can be extended to reweight MSMs by combining it with a reweighting of the Boltzmann distribution. We demonstrate how to efficiently implement the Girsanov reweighting in a molecular dynamics simulation program by calculating parts of the reweighting factor "on the fly" during the simulation, and we benchmark the method on test systems ranging from a two-dimensional diffusion process and an artificial many-body system to alanine dipeptide and valine dipeptide in implicit and explicit water. The method can be used to study the sensitivity of molecular dynamics on external perturbations as well as to reweight trajectories generated by enhanced sampling schemes to the original dynamics.

A comparison of large-scale climate signals and the North American Multi-Model Ensemble (NMME) for drought prediction in China

NASA Astrophysics Data System (ADS)

Xu, Lei; Chen, Nengcheng; Zhang, Xiang

2018-02-01

Drought is an extreme natural disaster that can lead to huge socioeconomic losses. Drought prediction ahead of months is helpful for early drought warning and preparations. In this study, we developed a statistical model, two weighted dynamic models and a statistical-dynamic (hybrid) model for 1-6 month lead drought prediction in China. Specifically, statistical component refers to climate signals weighting by support vector regression (SVR), dynamic components consist of the ensemble mean (EM) and Bayesian model averaging (BMA) of the North American Multi-Model Ensemble (NMME) climatic models, and the hybrid part denotes a combination of statistical and dynamic components by assigning weights based on their historical performances. The results indicate that the statistical and hybrid models show better rainfall predictions than NMME-EM and NMME-BMA models, which have good predictability only in southern China. In the 2011 China winter-spring drought event, the statistical model well predicted the spatial extent and severity of drought nationwide, although the severity was underestimated in the mid-lower reaches of Yangtze River (MLRYR) region. The NMME-EM and NMME-BMA models largely overestimated rainfall in northern and western China in 2011 drought. In the 2013 China summer drought, the NMME-EM model forecasted the drought extent and severity in eastern China well, while the statistical and hybrid models falsely detected negative precipitation anomaly (NPA) in some areas. Model ensembles such as multiple statistical approaches, multiple dynamic models or multiple hybrid models for drought predictions were highlighted. These conclusions may be helpful for drought prediction and early drought warnings in China.

Statistical Ensemble of Large Eddy Simulations

NASA Technical Reports Server (NTRS)

Carati, Daniele; Rogers, Michael M.; Wray, Alan A.; Mansour, Nagi N. (Technical Monitor)

2001-01-01

A statistical ensemble of large eddy simulations (LES) is run simultaneously for the same flow. The information provided by the different large scale velocity fields is used to propose an ensemble averaged version of the dynamic model. This produces local model parameters that only depend on the statistical properties of the flow. An important property of the ensemble averaged dynamic procedure is that it does not require any spatial averaging and can thus be used in fully inhomogeneous flows. Also, the ensemble of LES's provides statistics of the large scale velocity that can be used for building new models for the subgrid-scale stress tensor. The ensemble averaged dynamic procedure has been implemented with various models for three flows: decaying isotropic turbulence, forced isotropic turbulence, and the time developing plane wake. It is found that the results are almost independent of the number of LES's in the statistical ensemble provided that the ensemble contains at least 16 realizations.

On estimating attenuation from the amplitude of the spectrally whitened ambient seismic field

NASA Astrophysics Data System (ADS)

Weemstra, Cornelis; Westra, Willem; Snieder, Roel; Boschi, Lapo

2014-06-01

Measuring attenuation on the basis of interferometric, receiver-receiver surface waves is a non-trivial task: the amplitude, more than the phase, of ensemble-averaged cross-correlations is strongly affected by non-uniformities in the ambient wavefield. In addition, ambient noise data are typically pre-processed in ways that affect the amplitude itself. Some authors have recently attempted to measure attenuation in receiver-receiver cross-correlations obtained after the usual pre-processing of seismic ambient-noise records, including, most notably, spectral whitening. Spectral whitening replaces the cross-spectrum with a unit amplitude spectrum. It is generally assumed that cross-terms have cancelled each other prior to spectral whitening. Cross-terms are peaks in the cross-correlation due to simultaneously acting noise sources, that is, spurious traveltime delays due to constructive interference of signal coming from different sources. Cancellation of these cross-terms is a requirement for the successful retrieval of interferometric receiver-receiver signal and results from ensemble averaging. In practice, ensemble averaging is replaced by integrating over sufficiently long time or averaging over several cross-correlation windows. Contrary to the general assumption, we show in this study that cross-terms are not required to cancel each other prior to spectral whitening, but may also cancel each other after the whitening procedure. Specifically, we derive an analytic approximation for the amplitude difference associated with the reversed order of cancellation and normalization. Our approximation shows that an amplitude decrease results from the reversed order. This decrease is predominantly non-linear at small receiver-receiver distances: at distances smaller than approximately two wavelengths, whitening prior to ensemble averaging causes a significantly stronger decay of the cross-spectrum.

Statistical characterization of planar two-dimensional Rayleigh-Taylor mixing layers

NASA Astrophysics Data System (ADS)

Sendersky, Dmitry

2000-10-01

The statistical evolution of a planar, randomly perturbed fluid interface subject to Rayleigh-Taylor instability is explored through numerical simulation in two space dimensions. The data set, generated by the front-tracking code FronTier, is highly resolved and covers a large ensemble of initial perturbations, allowing a more refined analysis of closure issues pertinent to the stochastic modeling of chaotic fluid mixing. We closely approach a two-fold convergence of the mean two-phase flow: convergence of the numerical solution under computational mesh refinement, and statistical convergence under increasing ensemble size. Quantities that appear in the two-phase averaged Euler equations are computed directly and analyzed for numerical and statistical convergence. Bulk averages show a high degree of convergence, while interfacial averages are convergent only in the outer portions of the mixing zone, where there is a coherent array of bubble and spike tips. Comparison with the familiar bubble/spike penetration law h = alphaAgt 2 is complicated by the lack of scale invariance, inability to carry the simulations to late time, the increasing Mach numbers of the bubble/spike tips, and sensitivity to the method of data analysis. Finally, we use the simulation data to analyze some constitutive properties of the mixing process.

The Dropout Learning Algorithm

PubMed Central

Baldi, Pierre; Sadowski, Peter

2014-01-01

Dropout is a recently introduced algorithm for training neural network by randomly dropping units during training to prevent their co-adaptation. A mathematical analysis of some of the static and dynamic properties of dropout is provided using Bernoulli gating variables, general enough to accommodate dropout on units or connections, and with variable rates. The framework allows a complete analysis of the ensemble averaging properties of dropout in linear networks, which is useful to understand the non-linear case. The ensemble averaging properties of dropout in non-linear logistic networks result from three fundamental equations: (1) the approximation of the expectations of logistic functions by normalized geometric means, for which bounds and estimates are derived; (2) the algebraic equality between normalized geometric means of logistic functions with the logistic of the means, which mathematically characterizes logistic functions; and (3) the linearity of the means with respect to sums, as well as products of independent variables. The results are also extended to other classes of transfer functions, including rectified linear functions. Approximation errors tend to cancel each other and do not accumulate. Dropout can also be connected to stochastic neurons and used to predict firing rates, and to backpropagation by viewing the backward propagation as ensemble averaging in a dropout linear network. Moreover, the convergence properties of dropout can be understood in terms of stochastic gradient descent. Finally, for the regularization properties of dropout, the expectation of the dropout gradient is the gradient of the corresponding approximation ensemble, regularized by an adaptive weight decay term with a propensity for self-consistent variance minimization and sparse representations. PMID:24771879

Large ensemble modeling of last deglacial retreat of the West Antarctic Ice Sheet: comparison of simple and advanced statistical techniques

NASA Astrophysics Data System (ADS)

Pollard, D.; Chang, W.; Haran, M.; Applegate, P.; DeConto, R.

2015-11-01

A 3-D hybrid ice-sheet model is applied to the last deglacial retreat of the West Antarctic Ice Sheet over the last ~ 20 000 years. A large ensemble of 625 model runs is used to calibrate the model to modern and geologic data, including reconstructed grounding lines, relative sea-level records, elevation-age data and uplift rates, with an aggregate score computed for each run that measures overall model-data misfit. Two types of statistical methods are used to analyze the large-ensemble results: simple averaging weighted by the aggregate score, and more advanced Bayesian techniques involving Gaussian process-based emulation and calibration, and Markov chain Monte Carlo. Results for best-fit parameter ranges and envelopes of equivalent sea-level rise with the simple averaging method agree quite well with the more advanced techniques, but only for a large ensemble with full factorial parameter sampling. Best-fit parameter ranges confirm earlier values expected from prior model tuning, including large basal sliding coefficients on modern ocean beds. Each run is extended 5000 years into the "future" with idealized ramped climate warming. In the majority of runs with reasonable scores, this produces grounding-line retreat deep into the West Antarctic interior, and the analysis provides sea-level-rise envelopes with well defined parametric uncertainty bounds.

Interactive vs. Non-Interactive Multi-Model Ensembles

NASA Astrophysics Data System (ADS)

Duane, G. S.

2013-12-01

If the members of an ensemble of different models are allowed to interact with one another in run time, predictive skill can be improved as compared to that of any individual model or any average of indvidual model outputs. Inter-model connections in such an interactive ensemble can be trained, using historical data, so that the resulting ``supermodel' synchronizes with reality when used in weather-prediction mode, where the individual models perform data assimilation from each other (with trainable inter-model 'observation error') as well as from real observations. In climate-projection mode, parameters of the individual models are changed, as might occur from an increase in GHG levels, and one obtains relevant statistical properties of the new supermodel attractor. In simple cases, it has been shown that training of the inter-model connections with the old parameter values gives a supermodel that is still predictive when the parameter values are changed. Here we inquire as to the circumstances under which supermodel performance can be expected to exceed that of the customary weighted average of model outputs. We consider a supermodel formed from quasigeostrophic (QG) channel models with different forcing coefficients, and introduce an effective training scheme for the inter-model connections. We show that the blocked-zonal index cycle is reproduced better by the supermodel than by any non-interactive ensemble in the extreme case where the forcing coefficients of the different models are very large or very small. With realistic differences in forcing coefficients, as would be representative of actual differences among IPCC-class models, the usual linearity assumption is justified and a weighted average of model outputs is adequate. It is therefore hypothesized that supermodeling is likely to be useful in situations where there are qualitative model differences, as arising from sub-gridscale parameterizations, that affect overall model behavior. Otherwise the usual ex post facto averaging will probably suffice. The advantage of supermodeling is seen in statistics such as anticorrelation between blocking activity in the Atlantic and Pacific sectors, in the case of the QG channel model, rather than in overall blocking frequency. Likewise in climate models, the advantage of supermodeling is typically manifest in higher-order statistics rather than in quantities such as mean temperature.

Upper Limit of Weights in TAI Computation

NASA Technical Reports Server (NTRS)

Thomas, Claudine; Azoubib, Jacques

1996-01-01

The international reference time scale International Atomic Time (TAI) computed by the Bureau International des Poids et Mesures (BIPM) relies on a weighted average of data from a large number of atomic clocks. In it, the weight attributed to a given clock depends on its long-term stability. In this paper the TAI algorithm is used as the basis for a discussion of how to implement an upper limit of weight for clocks contributing to the ensemble time. This problem is approached through the comparison of two different techniques. In one case, a maximum relative weight is fixed: no individual clock can contribute more than a given fraction to the resulting time scale. The weight of each clock is then adjusted according to the qualities of the whole set of contributing elements. In the other case, a parameter characteristic of frequency stability is chosen: no individual clock can appear more stable than the stated limit. This is equivalent to choosing an absolute limit of weight and attributing this to to the most stable clocks independently of the other elements of the ensemble. The first technique is more robust than the second and automatically optimizes the stability of the resulting time scale, but leads to a more complicated computatio. The second technique has been used in the TAI algorithm since the very beginning. Careful analysis of tests on real clock data shows that improvement of the stability of the time scale requires revision from time to time of the fixed value chosen for the upper limit of absolute weight. In particular, we present results which confirm the decision of the CCDS Working Group on TAI to increase the absolute upper limit by a factor of 2.5. We also show that the use of an upper relative contribution further helps to improve the stability and may be a useful step towards better use of the massive ensemble of HP 507IA clocks now contributing to TAI.

Canonical-ensemble state-averaged complete active space self-consistent field (SA-CASSCF) strategy for problems with more diabatic than adiabatic states: Charge-bond resonance in monomethine cyanines

DOE Office of Scientific and Technical Information (OSTI.GOV)

Olsen, Seth, E-mail: seth.olsen@uq.edu.au

2015-01-28

This paper reviews basic results from a theory of the a priori classical probabilities (weights) in state-averaged complete active space self-consistent field (SA-CASSCF) models. It addresses how the classical probabilities limit the invariance of the self-consistency condition to transformations of the complete active space configuration interaction (CAS-CI) problem. Such transformations are of interest for choosing representations of the SA-CASSCF solution that are diabatic with respect to some interaction. I achieve the known result that a SA-CASSCF can be self-consistently transformed only within degenerate subspaces of the CAS-CI ensemble density matrix. For uniformly distributed (“microcanonical”) SA-CASSCF ensembles, self-consistency is invariant tomore » any unitary CAS-CI transformation that acts locally on the ensemble support. Most SA-CASSCF applications in current literature are microcanonical. A problem with microcanonical SA-CASSCF models for problems with “more diabatic than adiabatic” states is described. The problem is that not all diabatic energies and couplings are self-consistently resolvable. A canonical-ensemble SA-CASSCF strategy is proposed to solve the problem. For canonical-ensemble SA-CASSCF, the equilibrated ensemble is a Boltzmann density matrix parametrized by its own CAS-CI Hamiltonian and a Lagrange multiplier acting as an inverse “temperature,” unrelated to the physical temperature. Like the convergence criterion for microcanonical-ensemble SA-CASSCF, the equilibration condition for canonical-ensemble SA-CASSCF is invariant to transformations that act locally on the ensemble CAS-CI density matrix. The advantage of a canonical-ensemble description is that more adiabatic states can be included in the support of the ensemble without running into convergence problems. The constraint on the dimensionality of the problem is relieved by the introduction of an energy constraint. The method is illustrated with a complete active space valence-bond (CASVB) analysis of the charge/bond resonance electronic structure of a monomethine cyanine: Michler’s hydrol blue. The diabatic CASVB representation is shown to vary weakly for “temperatures” corresponding to visible photon energies. Canonical-ensemble SA-CASSCF enables the resolution of energies and couplings for all covalent and ionic CASVB structures contributing to the SA-CASSCF ensemble. The CASVB solution describes resonance of charge- and bond-localized electronic structures interacting via bridge resonance superexchange. The resonance couplings can be separated into channels associated with either covalent charge delocalization or chemical bonding interactions, with the latter significantly stronger than the former.« less

Canonical-ensemble state-averaged complete active space self-consistent field (SA-CASSCF) strategy for problems with more diabatic than adiabatic states: charge-bond resonance in monomethine cyanines.

PubMed

Olsen, Seth

2015-01-28

This paper reviews basic results from a theory of the a priori classical probabilities (weights) in state-averaged complete active space self-consistent field (SA-CASSCF) models. It addresses how the classical probabilities limit the invariance of the self-consistency condition to transformations of the complete active space configuration interaction (CAS-CI) problem. Such transformations are of interest for choosing representations of the SA-CASSCF solution that are diabatic with respect to some interaction. I achieve the known result that a SA-CASSCF can be self-consistently transformed only within degenerate subspaces of the CAS-CI ensemble density matrix. For uniformly distributed ("microcanonical") SA-CASSCF ensembles, self-consistency is invariant to any unitary CAS-CI transformation that acts locally on the ensemble support. Most SA-CASSCF applications in current literature are microcanonical. A problem with microcanonical SA-CASSCF models for problems with "more diabatic than adiabatic" states is described. The problem is that not all diabatic energies and couplings are self-consistently resolvable. A canonical-ensemble SA-CASSCF strategy is proposed to solve the problem. For canonical-ensemble SA-CASSCF, the equilibrated ensemble is a Boltzmann density matrix parametrized by its own CAS-CI Hamiltonian and a Lagrange multiplier acting as an inverse "temperature," unrelated to the physical temperature. Like the convergence criterion for microcanonical-ensemble SA-CASSCF, the equilibration condition for canonical-ensemble SA-CASSCF is invariant to transformations that act locally on the ensemble CAS-CI density matrix. The advantage of a canonical-ensemble description is that more adiabatic states can be included in the support of the ensemble without running into convergence problems. The constraint on the dimensionality of the problem is relieved by the introduction of an energy constraint. The method is illustrated with a complete active space valence-bond (CASVB) analysis of the charge/bond resonance electronic structure of a monomethine cyanine: Michler's hydrol blue. The diabatic CASVB representation is shown to vary weakly for "temperatures" corresponding to visible photon energies. Canonical-ensemble SA-CASSCF enables the resolution of energies and couplings for all covalent and ionic CASVB structures contributing to the SA-CASSCF ensemble. The CASVB solution describes resonance of charge- and bond-localized electronic structures interacting via bridge resonance superexchange. The resonance couplings can be separated into channels associated with either covalent charge delocalization or chemical bonding interactions, with the latter significantly stronger than the former.

Reducing the Uncertainty in Atlantic Meridional Overturning Circulation Projections Using Bayesian Model Averaging

NASA Astrophysics Data System (ADS)

Olson, R.; An, S. I.

2016-12-01

Atlantic Meridional Overturning Circulation (AMOC) in the ocean might slow down in the future, which can lead to a host of climatic effects in North Atlantic and throughout the world. Despite improvements in climate models and availability of new observations, AMOC projections remain uncertain. Here we constrain CMIP5 multi-model ensemble output with observations of a recently developed AMOC index to provide improved Bayesian predictions of future AMOC. Specifically, we first calculate yearly AMOC index loosely based on Rahmstorf et al. (2015) for years 1880—2004 for both observations, and the CMIP5 models for which relevant output is available. We then assign a weight to each model based on a Bayesian Model Averaging method that accounts for differential model skill in terms of both mean state and variability. We include the temporal autocorrelation in climate model errors, and account for the uncertainty in the parameters of our statistical model. We use the weights to provide future weighted projections of AMOC, and compare them to un-weighted ones. Our projections use bootstrapping to account for uncertainty in internal AMOC variability. We also perform spectral and other statistical analyses to show that AMOC index variability, both in models and in observations, is consistent with red noise. Our results improve on and complement previous work by using a new ensemble of climate models, a different observational metric, and an improved Bayesian weighting method that accounts for differential model skill at reproducing internal variability. Reference: Rahmstorf, S., Box, J. E., Feulner, G., Mann, M. E., Robinson, A., Rutherford, S., & Schaffernicht, E. J. (2015). Exceptional twentieth-century slowdown in atlantic ocean overturning circulation. Nature Climate Change, 5(5), 475-480. doi:10.1038/nclimate2554

Online probabilistic learning with an ensemble of forecasts

NASA Astrophysics Data System (ADS)

Thorey, Jean; Mallet, Vivien; Chaussin, Christophe

2016-04-01

Our objective is to produce a calibrated weighted ensemble to forecast a univariate time series. In addition to a meteorological ensemble of forecasts, we rely on observations or analyses of the target variable. The celebrated Continuous Ranked Probability Score (CRPS) is used to evaluate the probabilistic forecasts. However applying the CRPS on weighted empirical distribution functions (deriving from the weighted ensemble) may introduce a bias because of which minimizing the CRPS does not produce the optimal weights. Thus we propose an unbiased version of the CRPS which relies on clusters of members and is strictly proper. We adapt online learning methods for the minimization of the CRPS. These methods generate the weights associated to the members in the forecasted empirical distribution function. The weights are updated before each forecast step using only past observations and forecasts. Our learning algorithms provide the theoretical guarantee that, in the long run, the CRPS of the weighted forecasts is at least as good as the CRPS of any weighted ensemble with weights constant in time. In particular, the performance of our forecast is better than that of any subset ensemble with uniform weights. A noteworthy advantage of our algorithm is that it does not require any assumption on the distributions of the observations and forecasts, both for the application and for the theoretical guarantee to hold. As application example on meteorological forecasts for photovoltaic production integration, we show that our algorithm generates a calibrated probabilistic forecast, with significant performance improvements on probabilistic diagnostic tools (the CRPS, the reliability diagram and the rank histogram).

The public health impact of malaria vaccine RTS,S in malaria endemic Africa: country-specific predictions using 18 month follow-up Phase III data and simulation models.

PubMed

Penny, Melissa A; Galactionova, Katya; Tarantino, Michael; Tanner, Marcel; Smith, Thomas A

2015-07-29

The RTS,S/AS01 malaria vaccine candidate recently completed Phase III trials in 11 African sites. Recommendations for its deployment will partly depend on predictions of public health impact in endemic countries. Previous predictions of these used only limited information on underlying vaccine properties and have not considered country-specific contextual data. Each Phase III trial cohort was simulated explicitly using an ensemble of individual-based stochastic models, and many hypothetical vaccine profiles. The true profile was estimated by Bayesian fitting of these models to the site- and time-specific incidence of clinical malaria in both trial arms over 18 months of follow-up. Health impacts of implementation via two vaccine schedules in 43 endemic sub-Saharan African countries, using country-specific prevalence, access to care, immunisation coverage and demography data, were predicted via weighted averaging over many simulations. The efficacy against infection of three doses of vaccine was initially approximately 65 % (when immunising 6-12 week old infants) and 80 % (children 5-17 months old), with a 1 year half-life (exponential decay). Either schedule will avert substantial disease, but predicted impact strongly depends on the decay rate of vaccine effects and average transmission intensity. For the first time Phase III site- and time-specific data were available to estimate both the underlying profile of RTS,S/AS01 and likely country-specific health impacts. Initial efficacy will probably be high, but decay rapidly. Adding RTS,S to existing control programs, assuming continuation of current levels of malaria exposure and of health system performance, will potentially avert 100-580 malaria deaths and 45,000 to 80,000 clinical episodes per 100,000 fully vaccinated children over an initial 10-year phase.

Multi-Model Combination techniques for Hydrological Forecasting: Application to Distributed Model Intercomparison Project Results

DOE Office of Scientific and Technical Information (OSTI.GOV)

Ajami, N K; Duan, Q; Gao, X

2005-04-11

This paper examines several multi-model combination techniques: the Simple Multi-model Average (SMA), the Multi-Model Super Ensemble (MMSE), Modified Multi-Model Super Ensemble (M3SE) and the Weighted Average Method (WAM). These model combination techniques were evaluated using the results from the Distributed Model Intercomparison Project (DMIP), an international project sponsored by the National Weather Service (NWS) Office of Hydrologic Development (OHD). All of the multi-model combination results were obtained using uncalibrated DMIP model outputs and were compared against the best uncalibrated as well as the best calibrated individual model results. The purpose of this study is to understand how different combination techniquesmore » affect the skill levels of the multi-model predictions. This study revealed that the multi-model predictions obtained from uncalibrated single model predictions are generally better than any single member model predictions, even the best calibrated single model predictions. Furthermore, more sophisticated multi-model combination techniques that incorporated bias correction steps work better than simple multi-model average predictions or multi-model predictions without bias correction.« less

Evaluation of Multi-Model Ensemble System for Seasonal and Monthly Prediction

NASA Astrophysics Data System (ADS)

Zhang, Q.; Van den Dool, H. M.

2013-12-01

Since August 2011, the realtime seasonal forecasts of U.S. National Multi-Model Ensemble (NMME) have been made on 8th of each month by NCEP Climate Prediction Center (CPC). During the first year, the participating models were NCEP/CFSv1&2, GFDL/CM2.2, NCAR/U.Miami/COLA/CCSM3, NASA/GEOS5, IRI/ ECHAM-a & ECHAM-f for the realtime NMME forecast. The Canadian Meteorological Center CanCM3 and CM4 replaced the CFSv1 and IRI's models in the second year. The NMME team at CPC collects three variables, including precipitation, 2-meter temperature and sea surface temperature from each modeling center on a 1x1 global grid, removes systematic errors, makes the grand ensemble mean with equal weight for each model and constructs a probability forecast with equal weight for each member. The team then provides the NMME forecast to the operational CPC forecaster responsible for the seasonal and monthly outlook each month. Verification of the seasonal and monthly prediction from NMME is conducted by calculating the anomaly correlation (AC) from the 30-year hindcasts (1982-2011) of individual model and NMME ensemble. The motivation of this study is to provide skill benchmarks for future improvements of the NMME seasonal and monthly prediction system. The experimental (Phase I) stage of the project already supplies routine guidance to users of the NMME forecasts.

Locally Weighted Ensemble Clustering.

PubMed

Huang, Dong; Wang, Chang-Dong; Lai, Jian-Huang

2018-05-01

Due to its ability to combine multiple base clusterings into a probably better and more robust clustering, the ensemble clustering technique has been attracting increasing attention in recent years. Despite the significant success, one limitation to most of the existing ensemble clustering methods is that they generally treat all base clusterings equally regardless of their reliability, which makes them vulnerable to low-quality base clusterings. Although some efforts have been made to (globally) evaluate and weight the base clusterings, yet these methods tend to view each base clustering as an individual and neglect the local diversity of clusters inside the same base clustering. It remains an open problem how to evaluate the reliability of clusters and exploit the local diversity in the ensemble to enhance the consensus performance, especially, in the case when there is no access to data features or specific assumptions on data distribution. To address this, in this paper, we propose a novel ensemble clustering approach based on ensemble-driven cluster uncertainty estimation and local weighting strategy. In particular, the uncertainty of each cluster is estimated by considering the cluster labels in the entire ensemble via an entropic criterion. A novel ensemble-driven cluster validity measure is introduced, and a locally weighted co-association matrix is presented to serve as a summary for the ensemble of diverse clusters. With the local diversity in ensembles exploited, two novel consensus functions are further proposed. Extensive experiments on a variety of real-world datasets demonstrate the superiority of the proposed approach over the state-of-the-art.

A comparative analysis of 9 multi-model averaging approaches in hydrological continuous streamflow simulation

NASA Astrophysics Data System (ADS)

Arsenault, Richard; Gatien, Philippe; Renaud, Benoit; Brissette, François; Martel, Jean-Luc

2015-10-01

This study aims to test whether a weighted combination of several hydrological models can simulate flows more accurately than the models taken individually. In addition, the project attempts to identify the most efficient model averaging method and the optimal number of models to include in the weighting scheme. In order to address the first objective, streamflow was simulated using four lumped hydrological models (HSAMI, HMETS, MOHYSE and GR4J-6), each of which were calibrated with three different objective functions on 429 watersheds. The resulting 12 hydrographs (4 models × 3 metrics) were weighted and combined with the help of 9 averaging methods which are the simple arithmetic mean (SAM), Akaike information criterion (AICA), Bates-Granger (BGA), Bayes information criterion (BICA), Bayesian model averaging (BMA), Granger-Ramanathan average variant A, B and C (GRA, GRB and GRC) and the average by SCE-UA optimization (SCA). The same weights were then applied to the hydrographs in validation mode, and the Nash-Sutcliffe Efficiency metric was measured between the averaged and observed hydrographs. Statistical analyses were performed to compare the accuracy of weighted methods to that of individual models. A Kruskal-Wallis test and a multi-objective optimization algorithm were then used to identify the most efficient weighted method and the optimal number of models to integrate. Results suggest that the GRA, GRB, GRC and SCA weighted methods perform better than the individual members. Model averaging from these four methods were superior to the best of the individual members in 76% of the cases. Optimal combinations on all watersheds included at least one of each of the four hydrological models. None of the optimal combinations included all members of the ensemble of 12 hydrographs. The Granger-Ramanathan average variant C (GRC) is recommended as the best compromise between accuracy, speed of execution, and simplicity.

Scattering and extinction by spherical particles immersed in an absorbing host medium

NASA Astrophysics Data System (ADS)

Mishchenko, Michael I.; Dlugach, Janna M.

2018-05-01

Many applications of electromagnetic scattering involve particles immersed in an absorbing rather than lossless medium, thereby making the conventional scattering theory potentially inapplicable. To analyze this issue quantitatively, we employ the FORTRAN program developed recently on the basis of the first-principles electromagnetic theory to study far-field scattering by spherical particles embedded in an absorbing infinite host medium. We further examine the phenomenon of negative extinction identified recently for monodisperse spheres and uncover additional evidence in favor of its interference origin. We identify the main effects of increasing the width of the size distribution on the ensemble-averaged extinction efficiency factor and show that negative extinction can be eradicated by averaging over a very narrow size distribution. We also analyze, for the first time, the effects of absorption inside the host medium and ensemble averaging on the phase function and other elements of the Stokes scattering matrix. It is shown in particular that increasing absorption significantly suppresses the interference structure and can result in a dramatic expansion of the areas of positive polarization. Furthermore, the phase functions computed for larger effective size parameters can develop a very deep minimum at side-scattering angles bracketed by a strong diffraction peak in the forward direction and a pronounced backscattering maximum.

Global carbon assimilation system using a local ensemble Kalman filter with multiple ecosystem models

NASA Astrophysics Data System (ADS)

Zhang, Shupeng; Yi, Xue; Zheng, Xiaogu; Chen, Zhuoqi; Dan, Bo; Zhang, Xuanze

2014-11-01

In this paper, a global carbon assimilation system (GCAS) is developed for optimizing the global land surface carbon flux at 1° resolution using multiple ecosystem models. In GCAS, three ecosystem models, Boreal Ecosystem Productivity Simulator, Carnegie-Ames-Stanford Approach, and Community Atmosphere Biosphere Land Exchange, produce the prior fluxes, and an atmospheric transport model, Model for OZone And Related chemical Tracers, is used to calculate atmospheric CO2 concentrations resulting from these prior fluxes. A local ensemble Kalman filter is developed to assimilate atmospheric CO2 data observed at 92 stations to optimize the carbon flux for six land regions, and the Bayesian model averaging method is implemented in GCAS to calculate the weighted average of the optimized fluxes based on individual ecosystem models. The weights for the models are found according to the closeness of their forecasted CO2 concentration to observation. Results of this study show that the model weights vary in time and space, allowing for an optimum utilization of different strengths of different ecosystem models. It is also demonstrated that spatial localization is an effective technique to avoid spurious optimization results for regions that are not well constrained by the atmospheric data. Based on the multimodel optimized flux from GCAS, we found that the average global terrestrial carbon sink over the 2002-2008 period is 2.97 ± 1.1 PgC yr-1, and the sinks are 0.88 ± 0.52, 0.27 ± 0.33, 0.67 ± 0.39, 0.90 ± 0.68, 0.21 ± 0.31, and 0.04 ± 0.08 PgC yr-1 for the North America, South America, Africa, Eurasia, Tropical Asia, and Australia, respectively. This multimodel GCAS can be used to improve global carbon cycle estimation.

Simulation of tropical cyclone activity over the western North Pacific based on CMIP5 models

NASA Astrophysics Data System (ADS)

Shen, Haibo; Zhou, Weican; Zhao, Haikun

2017-09-01

Based on the Coupled Model Inter-comparison Project 5 (CMIP5) models, the tropical cyclone (TC) activity in the summers of 1965-2005 over the western North Pacific (WNP) is simulated by a TC dynamically downscaling system. In consideration of diversity among climate models, Bayesian model averaging (BMA) and equal-weighed model averaging (EMA) methods are applied to produce the ensemble large-scale environmental factors of the CMIP5 model outputs. The environmental factors generated by BMA and EMA methods are compared, as well as the corresponding TC simulations by the downscaling system. Results indicate that BMA method shows a significant advantage over the EMA. In addition, impacts of model selections on BMA method are examined. To each factor, ten models with better performance are selected from 30 CMIP5 models and then conduct BMA, respectively. As a consequence, the ensemble environmental factors and simulated TC activity are similar with the results from the 30 models' BMA, which verifies the BMA method can afford corresponding weight for each model in the ensemble based on the model's predictive skill. Thereby, the existence of poor performance models will not particularly affect the BMA effectiveness and the ensemble outcomes are improved. Finally, based upon the BMA method and downscaling system, we analyze the sensitivity of TC activity to three important environmental factors, i.e., sea surface temperature (SST), large-scale steering flow, and vertical wind shear. Among three factors, SST and large-scale steering flow greatly affect TC tracks, while average intensity distribution is sensitive to all three environmental factors. Moreover, SST and vertical wind shear jointly play a critical role in the inter-annual variability of TC lifetime maximum intensity and frequency of intense TCs.

A new transform for the analysis of complex fractionated atrial electrograms

PubMed Central

2011-01-01

Background Representation of independent biophysical sources using Fourier analysis can be inefficient because the basis is sinusoidal and general. When complex fractionated atrial electrograms (CFAE) are acquired during atrial fibrillation (AF), the electrogram morphology depends on the mix of distinct nonsinusoidal generators. Identification of these generators using efficient methods of representation and comparison would be useful for targeting catheter ablation sites to prevent arrhythmia reinduction. Method A data-driven basis and transform is described which utilizes the ensemble average of signal segments to identify and distinguish CFAE morphologic components and frequencies. Calculation of the dominant frequency (DF) of actual CFAE, and identification of simulated independent generator frequencies and morphologies embedded in CFAE, is done using a total of 216 recordings from 10 paroxysmal and 10 persistent AF patients. The transform is tested versus Fourier analysis to detect spectral components in the presence of phase noise and interference. Correspondence is shown between ensemble basis vectors of highest power and corresponding synthetic drivers embedded in CFAE. Results The ensemble basis is orthogonal, and efficient for representation of CFAE components as compared with Fourier analysis (p ≤ 0.002). When three synthetic drivers with additive phase noise and interference were decomposed, the top three peaks in the ensemble power spectrum corresponded to the driver frequencies more closely as compared with top Fourier power spectrum peaks (p ≤ 0.005). The synthesized drivers with phase noise and interference were extractable from their corresponding ensemble basis with a mean error of less than 10%. Conclusions The new transform is able to efficiently identify CFAE features using DF calculation and by discerning morphologic differences. Unlike the Fourier transform method, it does not distort CFAE signals prior to analysis, and is relatively robust to jitter in periodic events. Thus the ensemble method can provide a useful alternative for quantitative characterization of CFAE during clinical study. PMID:21569421

Constructing better classifier ensemble based on weighted accuracy and diversity measure.

PubMed

Zeng, Xiaodong; Wong, Derek F; Chao, Lidia S

2014-01-01

A weighted accuracy and diversity (WAD) method is presented, a novel measure used to evaluate the quality of the classifier ensemble, assisting in the ensemble selection task. The proposed measure is motivated by a commonly accepted hypothesis; that is, a robust classifier ensemble should not only be accurate but also different from every other member. In fact, accuracy and diversity are mutual restraint factors; that is, an ensemble with high accuracy may have low diversity, and an overly diverse ensemble may negatively affect accuracy. This study proposes a method to find the balance between accuracy and diversity that enhances the predictive ability of an ensemble for unknown data. The quality assessment for an ensemble is performed such that the final score is achieved by computing the harmonic mean of accuracy and diversity, where two weight parameters are used to balance them. The measure is compared to two representative measures, Kappa-Error and GenDiv, and two threshold measures that consider only accuracy or diversity, with two heuristic search algorithms, genetic algorithm, and forward hill-climbing algorithm, in ensemble selection tasks performed on 15 UCI benchmark datasets. The empirical results demonstrate that the WAD measure is superior to others in most cases.

Constructing Better Classifier Ensemble Based on Weighted Accuracy and Diversity Measure

PubMed Central

Chao, Lidia S.

2014-01-01

A weighted accuracy and diversity (WAD) method is presented, a novel measure used to evaluate the quality of the classifier ensemble, assisting in the ensemble selection task. The proposed measure is motivated by a commonly accepted hypothesis; that is, a robust classifier ensemble should not only be accurate but also different from every other member. In fact, accuracy and diversity are mutual restraint factors; that is, an ensemble with high accuracy may have low diversity, and an overly diverse ensemble may negatively affect accuracy. This study proposes a method to find the balance between accuracy and diversity that enhances the predictive ability of an ensemble for unknown data. The quality assessment for an ensemble is performed such that the final score is achieved by computing the harmonic mean of accuracy and diversity, where two weight parameters are used to balance them. The measure is compared to two representative measures, Kappa-Error and GenDiv, and two threshold measures that consider only accuracy or diversity, with two heuristic search algorithms, genetic algorithm, and forward hill-climbing algorithm, in ensemble selection tasks performed on 15 UCI benchmark datasets. The empirical results demonstrate that the WAD measure is superior to others in most cases. PMID:24672402

Multidimensional generalized-ensemble algorithms for complex systems.

PubMed

Mitsutake, Ayori; Okamoto, Yuko

2009-06-07

We give general formulations of the multidimensional multicanonical algorithm, simulated tempering, and replica-exchange method. We generalize the original potential energy function E(0) by adding any physical quantity V of interest as a new energy term. These multidimensional generalized-ensemble algorithms then perform a random walk not only in E(0) space but also in V space. Among the three algorithms, the replica-exchange method is the easiest to perform because the weight factor is just a product of regular Boltzmann-like factors, while the weight factors for the multicanonical algorithm and simulated tempering are not a priori known. We give a simple procedure for obtaining the weight factors for these two latter algorithms, which uses a short replica-exchange simulation and the multiple-histogram reweighting techniques. As an example of applications of these algorithms, we have performed a two-dimensional replica-exchange simulation and a two-dimensional simulated-tempering simulation using an alpha-helical peptide system. From these simulations, we study the helix-coil transitions of the peptide in gas phase and in aqueous solution.

Cosmological ensemble and directional averages of observables

DOE Office of Scientific and Technical Information (OSTI.GOV)

Bonvin, Camille; Clarkson, Chris; Durrer, Ruth

We show that at second order, ensemble averages of observables and directional averages do not commute due to gravitational lensing—observing the same thing in many directions over the sky is not the same as taking an ensemble average. In principle this non-commutativity is significant for a variety of quantities that we often use as observables and can lead to a bias in parameter estimation. We derive the relation between the ensemble average and the directional average of an observable, at second order in perturbation theory. We discuss the relevance of these two types of averages for making predictions of cosmologicalmore » observables, focusing on observables related to distances and magnitudes. In particular, we show that the ensemble average of the distance in a given observed direction is increased by gravitational lensing, whereas the directional average of the distance is decreased. For a generic observable, there exists a particular function of the observable that is not affected by second-order lensing perturbations. We also show that standard areas have an advantage over standard rulers, and we discuss the subtleties involved in averaging in the case of supernova observations.« less

A stochastic Markov chain model to describe lung cancer growth and metastasis.

PubMed

Newton, Paul K; Mason, Jeremy; Bethel, Kelly; Bazhenova, Lyudmila A; Nieva, Jorge; Kuhn, Peter

2012-01-01

A stochastic Markov chain model for metastatic progression is developed for primary lung cancer based on a network construction of metastatic sites with dynamics modeled as an ensemble of random walkers on the network. We calculate a transition matrix, with entries (transition probabilities) interpreted as random variables, and use it to construct a circular bi-directional network of primary and metastatic locations based on postmortem tissue analysis of 3827 autopsies on untreated patients documenting all primary tumor locations and metastatic sites from this population. The resulting 50 potential metastatic sites are connected by directed edges with distributed weightings, where the site connections and weightings are obtained by calculating the entries of an ensemble of transition matrices so that the steady-state distribution obtained from the long-time limit of the Markov chain dynamical system corresponds to the ensemble metastatic distribution obtained from the autopsy data set. We condition our search for a transition matrix on an initial distribution of metastatic tumors obtained from the data set. Through an iterative numerical search procedure, we adjust the entries of a sequence of approximations until a transition matrix with the correct steady-state is found (up to a numerical threshold). Since this constrained linear optimization problem is underdetermined, we characterize the statistical variance of the ensemble of transition matrices calculated using the means and variances of their singular value distributions as a diagnostic tool. We interpret the ensemble averaged transition probabilities as (approximately) normally distributed random variables. The model allows us to simulate and quantify disease progression pathways and timescales of progression from the lung position to other sites and we highlight several key findings based on the model.

Impact of climate change on precipitation distribution and water availability in the Nile using CMIP5 GCM ensemble.

NASA Astrophysics Data System (ADS)

Mekonnen, Z. T.; Gebremichael, M.

2017-12-01

ABSTRACT In a basin like the Nile where millions of people depend on rainfed agriculture and surface water resources for their livelihoods, changes in precipitation will have tremendous social and economic consequences. General circulation models (GCMs) have been associated with high uncertainty in their projection of future precipitation for the Nile basin. Some studies tried to compare performance of different GCMs by doing a Multi-Model comparison for the region. Many indicated that there is no single model that gives the "best estimate" of precipitation for a very complex and large basin like the Nile. In this study, we used a combination of satellite and long term rain gauge precipitation measurements (TRMM and CenTrends) to evaluate the performance of 10 GCMs from the 5th Coupled Model Intercomparison Project (CMIP5) at different spatial and seasonal scales and produce a weighted ensemble projection. Our results confirm that there is no single model that gives best estimate over the region, hence the approach of creating an ensemble depending on how the model performed in specific areas and seasons resulted in an improved estimate of precipitation compared with observed values. Following the same approach, we created an ensemble of future precipitation projections for four different time periods (2000-2024, 2025-2049 and 2050-2100). The analysis showed that all the major sub-basins of the Nile will get will get more precipitation with time, even though the distribution with in the sub basin might be different. Overall the analysis showed a 15 % increase (125 mm/year) by the end of the century averaged over the area up to the Aswan dam. KEY WORDS: Climate Change, CMIP5, Nile, East Africa, CenTrends, Precipitation, Weighted Ensembles

Ensemble predictive model for more accurate soil organic carbon spectroscopic estimation

NASA Astrophysics Data System (ADS)

Vašát, Radim; Kodešová, Radka; Borůvka, Luboš

2017-07-01

A myriad of signal pre-processing strategies and multivariate calibration techniques has been explored in attempt to improve the spectroscopic prediction of soil organic carbon (SOC) over the last few decades. Therefore, to come up with a novel, more powerful, and accurate predictive approach to beat the rank becomes a challenging task. However, there may be a way, so that combine several individual predictions into a single final one (according to ensemble learning theory). As this approach performs best when combining in nature different predictive algorithms that are calibrated with structurally different predictor variables, we tested predictors of two different kinds: 1) reflectance values (or transforms) at each wavelength and 2) absorption feature parameters. Consequently we applied four different calibration techniques, two per each type of predictors: a) partial least squares regression and support vector machines for type 1, and b) multiple linear regression and random forest for type 2. The weights to be assigned to individual predictions within the ensemble model (constructed as a weighted average) were determined by an automated procedure that ensured the best solution among all possible was selected. The approach was tested at soil samples taken from surface horizon of four sites differing in the prevailing soil units. By employing the ensemble predictive model the prediction accuracy of SOC improved at all four sites. The coefficient of determination in cross-validation (R2cv) increased from 0.849, 0.611, 0.811 and 0.644 (the best individual predictions) to 0.864, 0.650, 0.824 and 0.698 for Site 1, 2, 3 and 4, respectively. Generally, the ensemble model affected the final prediction so that the maximal deviations of predicted vs. observed values of the individual predictions were reduced, and thus the correlation cloud became thinner as desired.

Ensemble Pulsar Time Scale

NASA Astrophysics Data System (ADS)

Yin, Dong-shan; Gao, Yu-ping; Zhao, Shu-hong

2017-07-01

Millisecond pulsars can generate another type of time scale that is totally independent of the atomic time scale, because the physical mechanisms of the pulsar time scale and the atomic time scale are quite different from each other. Usually the pulsar timing observations are not evenly sampled, and the internals between two data points range from several hours to more than half a month. Further more, these data sets are sparse. All this makes it difficult to generate an ensemble pulsar time scale. Hence, a new algorithm to calculate the ensemble pulsar time scale is proposed. Firstly, a cubic spline interpolation is used to densify the data set, and make the intervals between data points uniform. Then, the Vondrak filter is employed to smooth the data set, and get rid of the high-frequency noises, and finally the weighted average method is adopted to generate the ensemble pulsar time scale. The newly released NANOGRAV (North American Nanohertz Observatory for Gravitational Waves) 9-year data set is used to generate the ensemble pulsar time scale. This data set includes the 9-year observational data of 37 millisecond pulsars observed by the 100-meter Green Bank telescope and the 305-meter Arecibo telescope. It is found that the algorithm used in this paper can reduce effectively the influence caused by the noises in pulsar timing residuals, and improve the long-term stability of the ensemble pulsar time scale. Results indicate that the long-term (> 1 yr) stability of the ensemble pulsar time scale is better than 3.4 × 10-15.

Quantitative precipitation forecasts in the Alps - an assessment from the Forecast Demonstration Project MAP D-PHASE

NASA Astrophysics Data System (ADS)

Ament, F.; Weusthoff, T.; Arpagaus, M.; Rotach, M.

2009-04-01

The main aim of the WWRP Forecast Demonstration Project MAP D-PHASE is to demonstrate the performance of today's models to forecast heavy precipitation and flood events in the Alpine region. Therefore an end-to-end, real-time forecasting system was installed and operated during the D PHASE Operations Period from June to November 2007. Part of this system are 30 numerical weather prediction models (deterministic as well as ensemble systems) operated by weather services and research institutes, which issue alerts if predicted precipitation accumulations exceed critical thresholds. Additionally to the real-time alerts, all relevant model fields of these simulations are stored in a central data archive. This comprehensive data set allows a detailed assessment of today's quantitative precipitation forecast (QPF) performance in the Alpine region. We will present results of QPF verifications against Swiss radar and rain gauge data both from a qualitative point of view, in terms of alerts, as well as from a quantitative perspective, in terms of precipitation rate. Various influencing factors like lead time, accumulation time, selection of warning thresholds, or bias corrections will be discussed. Additional to traditional verifications of area average precipitation amounts, the performance of the models to predict the correct precipitation statistics without requiring a point-to-point match will be described by using modern Fuzzy verification techniques. Both analyses reveal significant advantages of deep convection resolving models compared to coarser models with parameterized convection. An intercomparison of the model forecasts themselves reveals a remarkably high variability between different models, and makes it worthwhile to evaluate the potential of a multi-model ensemble. Various multi-model ensemble strategies will be tested by combining D-PHASE models to virtual ensemble systems.

A two-dimensional numerical study of the flow inside the combustion chambers of a motored rotary engine

NASA Technical Reports Server (NTRS)

Shih, T. I. P.; Yang, S. L.; Schock, H. J.

1986-01-01

A numerical study was performed to investigate the unsteady, multidimensional flow inside the combustion chambers of an idealized, two-dimensional, rotary engine under motored conditions. The numerical study was based on the time-dependent, two-dimensional, density-weighted, ensemble-averaged conservation equations of mass, species, momentum, and total energy valid for two-component ideal gas mixtures. The ensemble-averaged conservation equations were closed by a K-epsilon model of turbulence. This K-epsilon model of turbulence was modified to account for some of the effects of compressibility, streamline curvature, low-Reynolds number, and preferential stress dissipation. Numerical solutions to the conservation equations were obtained by the highly efficient implicit-factored method of Beam and Warming. The grid system needed to obtain solutions were generated by an algebraic grid generation technique based on transfinite interpolation. Results of the numerical study are presented in graphical form illustrating the flow patterns during intake, compression, gaseous fuel injection, expansion, and exhaust.

A two-dimensional numerical study of the flow inside the combustion chamber of a motored rotary engine

NASA Technical Reports Server (NTRS)

Shih, T. I-P.; Yang, S. L.; Schock, H. J.

1986-01-01

A numerical study was performed to investigate the unsteady, multidimensional flow inside the combustion chambers of an idealized, two-dimensional, rotary engine under motored conditions. The numerical study was based on the time-dependent, two-dimensional, density-weighted, ensemble-averaged conservation equations of mass, species, momentum, and total energy valid for two-component ideal gas mixtures. The ensemble-averaged conservation equations were closed by a K-epsilon model of turbulence. This K-epsilon model of turbulence was modified to account for some of the effects of compressibility, streamline curvature, low-Reynolds number, and preferential stress dissipation. Numerical solutions to the conservation equations were obtained by the highly efficient implicit-factored method of Beam and Warming. The grid system needed to obtain solutions were generated by an algebraic grid generation technique based on transfinite interpolation. Results of the numerical study are presented in graphical form illustrating the flow patterns during intake, compression, gaseous fuel injection, expansion, and exhaust.

Decadal climate predictions improved by ocean ensemble dispersion filtering

NASA Astrophysics Data System (ADS)

Kadow, C.; Illing, S.; Kröner, I.; Ulbrich, U.; Cubasch, U.

2017-06-01

Decadal predictions by Earth system models aim to capture the state and phase of the climate several years in advance. Atmosphere-ocean interaction plays an important role for such climate forecasts. While short-term weather forecasts represent an initial value problem and long-term climate projections represent a boundary condition problem, the decadal climate prediction falls in-between these two time scales. In recent years, more precise initialization techniques of coupled Earth system models and increased ensemble sizes have improved decadal predictions. However, climate models in general start losing the initialized signal and its predictive skill from one forecast year to the next. Here we show that the climate prediction skill of an Earth system model can be improved by a shift of the ocean state toward the ensemble mean of its individual members at seasonal intervals. We found that this procedure, called ensemble dispersion filter, results in more accurate results than the standard decadal prediction. Global mean and regional temperature, precipitation, and winter cyclone predictions show an increased skill up to 5 years ahead. Furthermore, the novel technique outperforms predictions with larger ensembles and higher resolution. Our results demonstrate how decadal climate predictions benefit from ocean ensemble dispersion filtering toward the ensemble mean.