Data-dependent posterior propriety of a Bayesian beta-binomial-logit model

Hyungsuk Tak, Carl N. Morris

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

A Beta-Binomial-Logit model is a Beta-Binomial model with covariate information incorporated via a logistic regression. Posterior propriety of a Bayesian Beta-Binomial-Logit model can be data-dependent for improper hyper-prior distributions. Various researchers in the literature have unknowingly used improper posterior distributions or have given incorrect statements about posterior propriety because checking posterior propriety can be challenging due to the complicated functional form of a Beta-Binomial-Logit model. We derive datadependent necessary and sufficient conditions for posterior propriety within a class of hyper-prior distributions that encompass those used in previous studies. When a posterior is improper due to improper hyper-prior distributions, we suggest using proper hyper-prior distributions that can mimic the behaviors of improper choices.

Original languageEnglish (US)
Pages (from-to)533-555
Number of pages23
JournalBayesian Analysis
Volume12
Issue number2
DOIs
StatePublished - Jun 1 2017

Fingerprint

Beta-binomial Model
Logit Model
Dependent Data
Prior distribution
Logistics
Logistic Regression
Posterior distribution
Covariates
Statistical Models
Necessary Conditions
Sufficient Conditions

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Applied Mathematics

Cite this

@article{35de8173adbe4843b038689a5a69912c,
title = "Data-dependent posterior propriety of a Bayesian beta-binomial-logit model",
abstract = "A Beta-Binomial-Logit model is a Beta-Binomial model with covariate information incorporated via a logistic regression. Posterior propriety of a Bayesian Beta-Binomial-Logit model can be data-dependent for improper hyper-prior distributions. Various researchers in the literature have unknowingly used improper posterior distributions or have given incorrect statements about posterior propriety because checking posterior propriety can be challenging due to the complicated functional form of a Beta-Binomial-Logit model. We derive datadependent necessary and sufficient conditions for posterior propriety within a class of hyper-prior distributions that encompass those used in previous studies. When a posterior is improper due to improper hyper-prior distributions, we suggest using proper hyper-prior distributions that can mimic the behaviors of improper choices.",
author = "Hyungsuk Tak and Morris, {Carl N.}",
year = "2017",
month = "6",
day = "1",
doi = "10.1214/16-BA1012",
language = "English (US)",
volume = "12",
pages = "533--555",
journal = "Bayesian Analysis",
issn = "1936-0975",
publisher = "Carnegie Mellon University",
number = "2",

}

Data-dependent posterior propriety of a Bayesian beta-binomial-logit model. / Tak, Hyungsuk; Morris, Carl N.

In: Bayesian Analysis, Vol. 12, No. 2, 01.06.2017, p. 533-555.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data-dependent posterior propriety of a Bayesian beta-binomial-logit model

AU - Tak, Hyungsuk

AU - Morris, Carl N.

PY - 2017/6/1

Y1 - 2017/6/1

N2 - A Beta-Binomial-Logit model is a Beta-Binomial model with covariate information incorporated via a logistic regression. Posterior propriety of a Bayesian Beta-Binomial-Logit model can be data-dependent for improper hyper-prior distributions. Various researchers in the literature have unknowingly used improper posterior distributions or have given incorrect statements about posterior propriety because checking posterior propriety can be challenging due to the complicated functional form of a Beta-Binomial-Logit model. We derive datadependent necessary and sufficient conditions for posterior propriety within a class of hyper-prior distributions that encompass those used in previous studies. When a posterior is improper due to improper hyper-prior distributions, we suggest using proper hyper-prior distributions that can mimic the behaviors of improper choices.

AB - A Beta-Binomial-Logit model is a Beta-Binomial model with covariate information incorporated via a logistic regression. Posterior propriety of a Bayesian Beta-Binomial-Logit model can be data-dependent for improper hyper-prior distributions. Various researchers in the literature have unknowingly used improper posterior distributions or have given incorrect statements about posterior propriety because checking posterior propriety can be challenging due to the complicated functional form of a Beta-Binomial-Logit model. We derive datadependent necessary and sufficient conditions for posterior propriety within a class of hyper-prior distributions that encompass those used in previous studies. When a posterior is improper due to improper hyper-prior distributions, we suggest using proper hyper-prior distributions that can mimic the behaviors of improper choices.

UR - http://www.scopus.com/inward/record.url?scp=85019159242&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85019159242&partnerID=8YFLogxK

U2 - 10.1214/16-BA1012

DO - 10.1214/16-BA1012

M3 - Article

AN - SCOPUS:85019159242

VL - 12

SP - 533

EP - 555

JO - Bayesian Analysis

JF - Bayesian Analysis

SN - 1936-0975

IS - 2

ER -