Online Program Home
My Program

Abstract Details

Activity Number: 439
Type: Contributed
Date/Time: Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #319682
Title: Bayesian Estimation with Flexible Prior for the Instrumental Variables Models
Author(s): John Hsu* and Julianne Swenson
Companies: University of California at Santa Barbara and University of California at Santa Barbara
Keywords: Inverse Wishart ; instrumental variables ; Markov Chain Monte Carlo ; Metropolis-Hastings algorithm

The instrumental variables method is a valuable tool in the analysis of simultaneous equations models. Since the estimation of coefficients in the model can be challenging, adept modeling of the covariance matrix is also important. The Inverse Wishart distribution is commonly used to provide a conjugate prior for the covariance matrix. However, the Inverse Wishart is limited in the flexibility to model prior information. We propose an alternative that allows for the specification of varying confidence levels for each element in the covariance matrix for the Instrumental Variables models. A real life example with corresponding results will be discussed in detail.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association