Prior distributions for variance parameters in hierarchical models

Andrew Gelman
Additional contact information
Andrew Gelman: Department of Statistics, Columbia University

Econometrics from University Library of Munich, Germany

Abstract: Various noninformative prior distributions have been suggested for scale parameters in hierarchical models. We construct a new folded-noncentral- t family of conditionally conjugate priors for hierarchical standard deviation parameters, and then consider noninformative and weakly informative priors in this family. We use an example to illustrate serious problems with the inverse-gamma family of ``noninformative'' prior distributions. We suggest instead to use a uniform prior on the hierarchical standard deviation, using the half-t family when the number of groups is small and in other settings where a weakly informative prior is desired.

Keywords: Bayesian inference; conditional conjugacy; folded noncentral-t distribution; half-t distribution; hierarchical model; multilevel model; noninformative prior distribution; weakly informative prior distribution (search for similar items in EconPapers)
JEL-codes: C1 C2 C3 C4 C5 C8 (search for similar items in EconPapers)
Pages: 13 pages
Date: 2004-04-14
New Economics Papers: this item is included in nep-ecm
Note: Type of Document - pdf; pages: 13
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (6)

Downloads: (external link)
https://econwpa.ub.uni-muenchen.de/econ-wp/em/papers/0404/0404001.pdf (application/pdf)

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:wpa:wuwpem:0404001

Access Statistics for this paper

More papers in Econometrics from University Library of Munich, Germany
Bibliographic data for series maintained by EconWPA (volker.schallehn@ub.uni-muenchen.de this e-mail address is bad, please contact repec@repec.org).