Online Program Home
My Program

Abstract Details

Activity Number: 102
Type: Invited
Date/Time: Monday, August 1, 2016 : 8:30 AM to 10:20 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #318428 View Presentation
Title: Nonparametric Regression with Instrumental Variables
Author(s): Purushottam Laud* and Robert McCulloch and Rodney Sparapani
Companies: Medical College of Wisconsin and The University of Chicago and Medical College of Wisconsin
Keywords: endogeneity ; non-linearity ; Gibbs conditionals

Regression models with instrumental variables are used widely in econometrics and in biomedical applications. Typically, a linear or additive model formulation is used with joint nonparametric errors. We employ Bayesian additive regression trees and a limited additivity assumption to allow much more flexible relationships that include interactions and non-linearities. After a full model description we address how Gibbs conditionals can be obtained for sampling the posterior. We illustrate the method with some well-known examples in the literature.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association