Online Program Home
My Program

Abstract Details

Activity Number: 268
Type: Contributed
Date/Time: Monday, August 1, 2016 : 2:00 PM to 2:45 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #321667
Title: Bayesian Model with Continuous Shrinkage Prior in Agricultural Health Study
Author(s): Ran Wei* and Subhashis Ghoshal and Brian J. Reich and Jane Hoppin
Companies: North Carolina State University and North Carolina State University and North Carolina State University and North Carolina State University
Keywords: Neurobehaviral ; Shrinkage ; Pesticide

This study is motivated by the Agricultural Health Study that focuses on the association between Neurobehavioral responses and pesticide exposures. The goal is to choose an appropriate subset of variables related to outcome measurements and evaluate the main effect as well interaction effect functions for all significant factors. We propose a Bayesian nonparametric regression model with Dirichlet-Laplace for variable selection and model prediction. This regression model first reduces the nonparametric regression function to additive linear settings by basis expansion in a way that each main effect and interaction effect is associated with a basis function and coefficient vector. The continuous shrinkage priors are imposed to shrink the coefficients of trivial variables. With this model, the individual function between responses and nontrivial chemical exposures can be estimated under specific restrictions. Theoretical proofs are followed to study the asymptotic behaviors of Dirichlet-Laplace prior for linear regression model.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association