Asymptotics of the expected posterior

Bertrand S Clarke, Dongchu Sun

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Let (X1, . . . , Xn) be independently and identically distributed observations from an exponential family pθ equipped with a smooth prior density w on a real d-dimensional parameter θ. We give conditions under which the expected value of the posterior density evaluated at the true value of the parameter, θ0, admits an asymptotic expansion in terms of the Fisher information I(θ0), the prior w, and their first two derivatives. The leading term of the expansion is of the form nd/2C1(d, θ0) and the second order term is of the form nd/2-1c2(d, θ0, w), with an error term that is o(nd/2-1). We identify the functions C1 and C2 explicitly. A modification of the proof of this expansion gives an analogous result for the expectation of the square of the posterior evaluated at θ0. As a consequence we can give a confidence band for the expected posterior, and we suggest a frequentist refinement for Bayesian testing.

Original languageEnglish (US)
Pages (from-to)163-185
Number of pages23
JournalAnnals of the Institute of Statistical Mathematics
Volume51
Issue number1
DOIs
StatePublished - Jan 1 1999

Fingerprint

Confidence Bands
Fisher Information
Exponential Family
Term
Error term
Expected Value
Identically distributed
Asymptotic Expansion
Refinement
Derivative
Testing
Form
Observation

Keywords

  • Asymptotics
  • Bayes factor
  • Chi-squared distance
  • Expected posterior
  • Relative entropy

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Asymptotics of the expected posterior. / Clarke, Bertrand S; Sun, Dongchu.

In: Annals of the Institute of Statistical Mathematics, Vol. 51, No. 1, 01.01.1999, p. 163-185.

Research output: Contribution to journalArticle

Clarke, Bertrand S ; Sun, Dongchu. / Asymptotics of the expected posterior. In: Annals of the Institute of Statistical Mathematics. 1999 ; Vol. 51, No. 1. pp. 163-185.
@article{26d1589ddc0342a5927944629e362644,
title = "Asymptotics of the expected posterior",
abstract = "Let (X1, . . . , Xn) be independently and identically distributed observations from an exponential family pθ equipped with a smooth prior density w on a real d-dimensional parameter θ. We give conditions under which the expected value of the posterior density evaluated at the true value of the parameter, θ0, admits an asymptotic expansion in terms of the Fisher information I(θ0), the prior w, and their first two derivatives. The leading term of the expansion is of the form nd/2C1(d, θ0) and the second order term is of the form nd/2-1c2(d, θ0, w), with an error term that is o(nd/2-1). We identify the functions C1 and C2 explicitly. A modification of the proof of this expansion gives an analogous result for the expectation of the square of the posterior evaluated at θ0. As a consequence we can give a confidence band for the expected posterior, and we suggest a frequentist refinement for Bayesian testing.",
keywords = "Asymptotics, Bayes factor, Chi-squared distance, Expected posterior, Relative entropy",
author = "Clarke, {Bertrand S} and Dongchu Sun",
year = "1999",
month = "1",
day = "1",
doi = "10.1023/A:1003891404142",
language = "English (US)",
volume = "51",
pages = "163--185",
journal = "Annals of the Institute of Statistical Mathematics",
issn = "0020-3157",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Asymptotics of the expected posterior

AU - Clarke, Bertrand S

AU - Sun, Dongchu

PY - 1999/1/1

Y1 - 1999/1/1

N2 - Let (X1, . . . , Xn) be independently and identically distributed observations from an exponential family pθ equipped with a smooth prior density w on a real d-dimensional parameter θ. We give conditions under which the expected value of the posterior density evaluated at the true value of the parameter, θ0, admits an asymptotic expansion in terms of the Fisher information I(θ0), the prior w, and their first two derivatives. The leading term of the expansion is of the form nd/2C1(d, θ0) and the second order term is of the form nd/2-1c2(d, θ0, w), with an error term that is o(nd/2-1). We identify the functions C1 and C2 explicitly. A modification of the proof of this expansion gives an analogous result for the expectation of the square of the posterior evaluated at θ0. As a consequence we can give a confidence band for the expected posterior, and we suggest a frequentist refinement for Bayesian testing.

AB - Let (X1, . . . , Xn) be independently and identically distributed observations from an exponential family pθ equipped with a smooth prior density w on a real d-dimensional parameter θ. We give conditions under which the expected value of the posterior density evaluated at the true value of the parameter, θ0, admits an asymptotic expansion in terms of the Fisher information I(θ0), the prior w, and their first two derivatives. The leading term of the expansion is of the form nd/2C1(d, θ0) and the second order term is of the form nd/2-1c2(d, θ0, w), with an error term that is o(nd/2-1). We identify the functions C1 and C2 explicitly. A modification of the proof of this expansion gives an analogous result for the expectation of the square of the posterior evaluated at θ0. As a consequence we can give a confidence band for the expected posterior, and we suggest a frequentist refinement for Bayesian testing.

KW - Asymptotics

KW - Bayes factor

KW - Chi-squared distance

KW - Expected posterior

KW - Relative entropy

UR - http://www.scopus.com/inward/record.url?scp=6544265019&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=6544265019&partnerID=8YFLogxK

U2 - 10.1023/A:1003891404142

DO - 10.1023/A:1003891404142

M3 - Article

VL - 51

SP - 163

EP - 185

JO - Annals of the Institute of Statistical Mathematics

JF - Annals of the Institute of Statistical Mathematics

SN - 0020-3157

IS - 1

ER -