Asymptotically Informative Prior for Bayesian Analysis

Ao Yuan and Jan G. Gooijer
Additional contact information
Ao Yuan: Howard University, Washington DC, USA

No 11-130/4, Tinbergen Institute Discussion Papers from Tinbergen Institute

Abstract: In classical Bayesian inference the prior is treated as fixed, it is asymptotically negligible,thus any information contained in the prior is ignored from the asymptotic first order result.However, in practice often an informative prior is summarized from previous similar or the samekind of studies, which contains non-negligible information for the current study. Here, differentfrom traditional Bayesian point of view, we treat such prior to be non-fixed. In particular,we give the data sizes used in previous studies for the prior the same status as the size of thecurrent dataset, viewing both sample sizes as increasing to infinity in the asymptotic study.Thus the prior is asymptotically non-negligible, and its original effects are ressumed under thisview. Consequently, Bayesian inference using such prior is more efficient, as it should be, thanthat regarded under the existing setting. We study some basic properties of Bayesian estimatorsusing such priors under convex losses and the 0—1 loss, and illustrate the method by an examplevia simulation.

Keywords: Asymptotically informative prior; asymptotic efficiency; Bayes estimator; information bound; maximum likelihood estimator (search for similar items in EconPapers)
JEL-codes: C10 C12 (search for similar items in EconPapers)
Date: 2011-09-15
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://papers.tinbergen.nl/11130.pdf (application/pdf)

Related works:
Journal Article: Asymptotically Informative Prior for Bayesian Analysis (2014)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:tin:wpaper:20110130

Access Statistics for this paper

More papers in Tinbergen Institute Discussion Papers from Tinbergen Institute Contact information at EDIRC.
Bibliographic data for series maintained by Tinbergen Office +31 (0)10-4088900 ().