Reducing biases in regional climate downscaling by applying Bayesian model averaging on large-scale forcing

2011 ◽  
Vol 39 (9-10) ◽  
pp. 2523-2532 ◽  
Author(s):  
Hongwei Yang ◽  
Bin Wang ◽  
Bin Wang
Keyword(s):  
Bayesian Model ◽  
Large Scale ◽  
Bayesian Model Averaging ◽  
Regional Climate ◽  
Model Averaging ◽  
Climate Downscaling

Optimizing Prediction Using Bayesian Model Averaging: Examples Using Large-Scale Educational Assessments

2018 ◽  
Vol 42 (4) ◽  
pp. 423-457 ◽  
Author(s):  
David Kaplan ◽  
Chansoon Lee
Keyword(s):  
Bayesian Model ◽  
Structural Equation ◽  
Large Scale ◽  
Bayesian Model Averaging ◽  
Model Averaging ◽  
Predictive Performance ◽  
Bayesian Regression ◽  
Equation Modeling ◽  
Scientific Principles ◽  
Educational Assessments

This article provides a review of Bayesian model averaging as a means of optimizing the predictive performance of common statistical models applied to large-scale educational assessments. The Bayesian framework recognizes that in addition to parameter uncertainty, there is uncertainty in the choice of models themselves. A Bayesian approach to addressing the problem of model uncertainty is the method of Bayesian model averaging. Bayesian model averaging searches the space of possible models for a set of submodels that satisfy certain scientific principles and then averages the coefficients across these submodels weighted by each model’s posterior model probability (PMP). Using the weighted coefficients for prediction has been shown to yield optimal predictive performance according to certain scoring rules. We demonstrate the utility of Bayesian model averaging for prediction in education research with three examples: Bayesian regression analysis, Bayesian logistic regression, and a recently developed approach for Bayesian structural equation modeling. In each case, the model-averaged estimates are shown to yield better prediction of the outcome of interest than any submodel based on predictive coverage and the log-score rule. Implications for the design of large-scale assessments when the goal is optimal prediction in a policy context are discussed.

scholarly journals Spatially-dependent Bayesian model selection for disease mapping

2016 ◽  
Vol 27 (1) ◽  
pp. 250-268 ◽  
Author(s):  
Rachel Carroll ◽  
Andrew B Lawson ◽  
Christel Faes ◽  
Russell S Kirby ◽  
Mehreteab Aregay ◽  
...  
Keyword(s):  
Model Selection ◽  
Bayesian Model ◽  
Large Scale ◽  
Bayesian Model Averaging ◽  
Disease Mapping ◽  
Model Averaging ◽  
Bayesian Model Selection ◽  
Small Scale ◽  
Study Region ◽  
Model Sets

In disease mapping where predictor effects are to be modeled, it is often the case that sets of predictors are fixed, and the aim is to choose between fixed model sets. Model selection methods, both Bayesian model selection and Bayesian model averaging, are approaches within the Bayesian paradigm for achieving this aim. In the spatial context, model selection could have a spatial component in the sense that some models may be more appropriate for certain areas of a study region than others. In this work, we examine the use of spatially referenced Bayesian model averaging and Bayesian model selection via a large-scale simulation study accompanied by a small-scale case study. Our results suggest that BMS performs well when a strong regression signature is found.

scholarly journals Bayesian model averaging of the RegCM temperature projections: a Canadian case study

2021 ◽  
Author(s):  
Tangnyu Song ◽  
Guohe Huang ◽  
Guoqing Wang ◽  
Yongping Li ◽  
Xiuquan Wang ◽  
...  
Keyword(s):  
Bayesian Model ◽  
General Circulation ◽  
Bayesian Model Averaging ◽  
Climate Model ◽  
Regional Climate ◽  
Model Averaging ◽  
Circulation Model ◽  
Ensemble Member ◽  
Eastern Canada ◽  
Pbl Scheme

Abstract The choices of physical schemes coupled in the regional climate model (RegCM), the input general circulation model (GCM) results, and the emission scenarios may cause considerable uncertainties in future temperature projections. Therefore, the ensemble approach, which can be used to reflect these uncertainties, is highly desired. In this study, the probabilistic projections for future temperature are generated at 88 Canadian climate stations based on the developed RegCM ensemble and obtained Bayesian model averaging (BMA) weights. The BMA weights indicate that the RegCM coupled with the holtslag PBL scheme driven by the HadGEM can provide relatively reliable temperature projections at most climate stations. It is also suggested that the BMA approach is effective in simulating temperature over middle and eastern Canada through taking the advantage of each ensemble member. However, the effectiveness of the BMA method is limited when all the models in the ensemble cannot simulate the temperature robustly. The projected results demonstrate that the temperature will increase continuously in the future, while the temperature increase under RCP8.5 will be significantly larger than that under RCP4.5.

scholarly journals Projected Climate Change Impacts on Future Streamflow of the Yarlung Tsangpo-Brahmaputra River

2018 ◽  
Author(s):  
Ran Xu ◽  
Hongchang Hu ◽  
Fuqiang Tian ◽  
Chao Li ◽  
Mohd Yawar Ali Khan
Keyword(s):  
Climate Change ◽  
Bias Correction ◽  
Bayesian Model ◽  
Bayesian Model Averaging ◽  
Regional Climate ◽  
Model Averaging ◽  
The Tibetan Plateau ◽  
Brahmaputra River ◽  
Yarlung Tsangpo ◽  
Impacts Of Climate Change

Abstract. The Yarlung Tsangpo-Brahmaputra River (YBR) originating from the Tibetan Plateau (TP), is an important water source for many domestic and agricultural practices in countries including China, India, Bhutan and Bangladesh. To date, only a few studies have investigated the impacts of climate change on water resources in this river basin with dispersed results. In this study, we provide a comprehensive and updated assessment of the impacts of climate change on YBR streamflow by integrating a physically based hydrological model, regional climate integrations from CORDEX (Coordinated Regional Climate Downscaling Experiment), different bias correction methods, and Bayesian model averaging method. We find that (i) bias correction is able to reduce systematic biases in regional climate integrations and thus benefits hydrological simulations over YBR Basin; (ii) Bayesian model averaging, which optimally combines individual hydrological simulations obtained from different bias correction methods, tends to provide hydrological time series superior over individual ones. We show that by the year 2035, the annual mean streamflow is projected to change respectively by 6.8 %, −0.4 %, and −4.1 % under RCP4.5 relative to the historical period (1980–2001) at the Bahadurabad in Bangladesh, the upper Brahmaputra outlet, and Nuxia in China. Under RCP8.5, these percentage changes will substantially increase to 12.9 %, 13.1 %, and 19.9 %. Therefore, the change rate of streamflow shows strong spatial variability along the YBR from downstream to upstream. The increasing rate of streamflow shows an augmented trend from downstream to upstream under RCP8.5 compared to an attenuated pattern under RCP4.5.

Selecting Predictors by Using Bayesian Model Averaging in Bridge Models

2012 ◽  
Author(s):  
Lorenzo Bencivelli ◽  
Massimiliano Giuseppe Marcellino ◽  
Gianluca Moretti
Keyword(s):  
Bayesian Model ◽  
Bayesian Model Averaging ◽  
Model Averaging ◽  
Bridge Models
Sign in / Sign up

Export Citation Format

Share Document