Online Program Home
My Program

Abstract Details

Activity Number: 427 - SPEED: Bayesian Methods, Part 2
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 3:05 PM to 3:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #307868
Title: An Investigation into How Model Uncertainty Is Reflected Through the Posterior Variance for Partial Regression Coefficients
Author(s): Katharine Banner* and Megan Higgs
Companies: Montana State University and Montana State University
Keywords: model selection; all subsets regression; avoiding naive inferences

Statistical inferences based on a single model are criticized for ignoring model uncertainty and overselling faith in model results. Accounting for model uncertainty with Bayesian model averaging (BMA) makes explanatory inference challenging and it may be unclear how a resulting posterior variance captures model uncertainty. An understanding of changes in posterior variances of partial regression coefficients across approaches will help inform choices about when to use BMA. Within the context of all subsets regression, we investigate three strategies: full model averaging, using the entire model set with posterior model probabilities (PMP) as weights; model averaging over a subset of models that include a particular covariate of interest, using re-normalized PMP; and making inference from one model with the largest PMP. For each strategy, we consider different correlations among covariates and signal-to-noise ratios in the data-generating model, and whether or not the data-generating model is in the model set. Counterintuitively, the posterior variance for a model-averaged partial regression coefficient is not always larger when considering more models.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program