• français
    • English
  • English 
    • français
    • English
  • Login
JavaScript is disabled for your browser. Some features of this site may not work without it.
BIRD Home

Browse

This CollectionBy Issue DateAuthorsTitlesSubjectsJournals BIRDResearch centres & CollectionsBy Issue DateAuthorsTitlesSubjectsJournals

My Account

Login

Statistics

View Usage Statistics

Model Misspecification in ABC: Consequences and Diagnostics

Thumbnail
View/Open
1708.01974.pdf (333.2Kb)
Date
2020
Dewey
Probabilités et mathématiques appliquées
Sujet
approximate Bayesian computation; Asymptotics; Likelihood‐free methods; Model misspecification; Regression adjustment approximate Bayesian computation
Journal issue
Journal of the Royal Statistical Society. Series B, Statistical Methodology
Volume
82
Number
2
Publication date
2020
Article pages
421-444
Publisher
Wiley
DOI
http://dx.doi.org/10.1111/rssb.12356
Forthcoming
oui
URI
https://basepub.dauphine.fr/handle/123456789/20365
Collections
  • CEREMADE : Publications
Metadata
Show full item record
Author
Frazier, David T.
231820 Monash University, Department of Economics
Robert, Christian P.
60 CEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
Rousseau, Judith
60 CEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
Type
Article accepté pour publication ou publié
Abstract (EN)
We analyze the behavior of approximate Bayesian computation (ABC) when the model generating the simulated data differs from the actual data generating process; i.e., when the data simulator in ABC is misspecified. We demonstrate both theoretically and in simple, but practically relevant, examples that when the model is misspecified different versions of ABC can lead to substantially different results. Our theoretical results demonstrate that under regularity conditions a version of the accept/reject ABC approach concentrates posterior mass on an appropriately defined pseudo-true parameter value. However, under model misspecification the ABC posterior does not yield credible sets with valid frequentist coverage and has non-standard asymptotic behavior. We also examine the theoretical behavior of the popular linear regression adjustment to ABC under model misspecification and demonstrate that this approach concentrates posterior mass on a completely different pseudo-true value than that obtained by the accept/reject approach to ABC. Using our theoretical results, we suggest two approaches to diagnose model misspecification in ABC. All theoretical results and diagnostics are illustrated in a simple running example.

  • Accueil Bibliothèque
  • Site de l'Université Paris-Dauphine
  • Contact
SCD Paris Dauphine - Place du Maréchal de Lattre de Tassigny 75775 Paris Cedex 16

 Content on this site is licensed under a Creative Commons 2.0 France (CC BY-NC-ND 2.0) license.