Online Program Home
My Program

Abstract Details

Activity Number: 302 - Advances in Bayesian Computation
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 8:30 AM to 10:20 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #307100
Title: Bayesian Assurance and Sample Size Analysis in a Conjugate Bayesian Linear Model Framework
Author(s): Jane Pan* and Sudipto Banerjee
Companies: UCLA and UCLA
Keywords: Bayesian Assurance; Conjugate Linear Models; Sample Size

Bayesian approaches for sample size analysis and power have proceeded from different perspectives. Recent developments on the effective use of design and analysis priors appear to be gaining ground and becoming more prevalent among various applications. We elucidate the use of such approaches within a conjugate Bayesian linear model framework and present situations in which Bayesian assurance coincides with frequentist power as well as showcase the advantages of approaching from a Bayesian standpoint. Our work aims to create open-source user-friendly data tools that allow easy access for statisticians to conduct Bayesian sample size analysis. We present a novel Bayesian sample size and assurance package and translate existing Bayesian modeling tools into user-interactive environments via the Shiny tool in R for a variety of Bayesian design problems within the linear regression context.

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

Back to the full JSM 2019 program