Online Program

Return to main conference page
Friday, October 4
Fri, Oct 4, 5:15 PM - 6:30 PM
Evergreen Ballroom Prefunction
Celebrating Women in Statistics and Data Science Reception and Poster Session 3

A Bayesian Approach to Quantile Envelope Regression (306723)

Saptarshi Chakraborty, Memorial Sloan Kettering Cancer Center 
*Minji Lee, University of Florida 
Zhihua Su, University of Florida 

Keywords: quantile regression, dimension reduction, Bayesian approach, envelope model

We propose a Bayesian quantile envelope model that adapts a nascent construct called envelope in Bayesian perspective. The Bayesian quantile envelope model can achieve the efficiency gains compared to the standard Bayesian quantile model. We provide a simple block Metropolis-within-Gibbs MCMC sampler for applications. We also demonstrate that our method performs well through simulations and data analysis.