We conduct both an approximate Bayesian Model Averaging (BMA) and an exact Bayesian analysis to incorporate break date uncertainty of the mean growth rate into the trend-cycle decomposition of U.S. real GDP. Our results suggest a structural break in mean growth rate of U.S. real GDP in 1970s. Comparing to the models assuming fixed break date, we find higher uncertainty in the posterior density ...

The Bayesian model averaging (BMA) method has been widely used for generating probabilistic climate projections. However, the weights in BMA can only reflect spatially- and temporally-averaged performance of each ensemble member, without ability to address spatiotemporal variations biases. This lead inevitable exaggeration or understatement contributions individual members mean, thus reducing r...

Temporal difference (TD) algorithms are attractive for reinforcement learning due to their ease-of-implementation and use of “bootstrapped” return estimates to make efficient use of sampled data. In particular, TD(λ) methods comprise a family of reinforcement learning algorithms that often yield fast convergence by averaging multiple estimators of the expected return. However, TD(λ) chooses a v...

We study the out-of-sample forecast performance of two alternative methods for dealing with dimensionality: Bayesian model Averaging (BMA) and principal components regression (PCR). We conduct a different out-of-sample investigation in which the predictors are chosen jointly for both output and inflation using Bayesian variable selection in each out-of-sample recursion using information availab...

This manual is a brief introduction to applied Bayesian Model Averaging with the R package BMS. The manual is structured as a hands-on tutorial for readers with few experience with BMA. Readers from a more technical background are advised to consult the table of contents for formal representations of the concepts used in BMS. For other tutorials and more information, please refer to http://bms....

We assess the accuracy of Bayesian polynomial extrapolations from small parameter values, x, to large values x. consider a set polynomials fixed order, intended as proxy for fixed-order effective field theory (EFT) description data. employ Model Averaging (BMA) combine results different order (EFT orders). Our study considers two "toy problems" where underlying function used generate data sets ...

[1] Most studies in vadose zone hydrology use a single conceptual model for predictive inference and analysis. Focusing on the outcome of a single model is prone to statistical bias and underestimation of uncertainty. In this study, we combine multiobjective optimization and Bayesian model averaging (BMA) to generate forecast ensembles of soil hydraulic models. To illustrate our method, we use ...

ABSTRACT This study considered the problem of predicting survival, based on three alternative models: a single Weibull, a mixture of Weibulls and a cure model. Instead of the common procedure of choosing a single "best" model, where "best" is defined in terms of goodness of fit to the data, a Bayesian model averaging (BMA) approach was adopted to account for model uncertainty. This was illustra...

Consideration of multiple models is routine statistical practice. With computational advances over the past decade, there has been increased interest in methods for making inferences based on combining models. Examples include boosting, bagging, stacking, and Bayesian Model Averaging (BMA), which often lead to improved performance over methods based on selecting a single model. Bernardo and Smi...

Bayesian model averaging (BMA) is a statistical postprocessing technique that generates calibrated and sharp predictive probability density functions (PDFs) from forecast ensembles. It represents the predictive PDF as a weighted average of PDFs centered on the bias-corrected ensemble members, where the weights reflect the relative skill of the individual members over a training period. This wor...