On the nature and practical handling of the Bayesian aggregation anomaly
Sergio Guarro and
Michael Yau
Reliability Engineering and System Safety, 2009, vol. 94, issue 6, 1050-1056
Abstract:
This paper addresses the issue, arising in probabilistic parameter estimation, which is sometimes referred to as the “Bayesian anomaly,†or also as the problem of “imperfect aggregation in Bayesian estimation.†The issue is clarified here from a generalized and theoretical point of view, but also in terms of a practical and mathematically consistent solution for probabilistic parameter estimation cases arising in actual reliability analysis and probabilistic risk assessment applications. More specifically, both theoretical and practical technical arguments are presented to show the reasons that make the purported anomaly manifest itself, and how the situations where it could in theory lead to serious estimation errors can be correctly handled. Interpretations and conclusions previously drawn with respect to the anomaly are also examined and re-evaluated in light of the new information and developments presented by the paper.
Keywords: Bayesian estimation; System reliability; Launch vehicle risk assessment; Bayesian anomaly; Aggregation anomaly (search for similar items in EconPapers)
Date: 2009
References: View complete reference list from CitEc
Citations: View citations in EconPapers (1)
Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0951832008002755
Full text for ScienceDirect subscribers only
Related works:
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:eee:reensy:v:94:y:2009:i:6:p:1050-1056
DOI: 10.1016/j.ress.2008.11.009
Access Statistics for this article
Reliability Engineering and System Safety is currently edited by Carlos Guedes Soares
More articles in Reliability Engineering and System Safety from Elsevier
Bibliographic data for series maintained by Catherine Liu ().