Marginalization Paradoxes in Bayesian and Structural Inference
Authors:
- 1 January 1973
- journal article
- research article
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series B: Statistical Methodology
- Vol. 35 (2) , 189-213
Abstract
Summary: We describe a range of routine statistical problems in which marginal posterior distributions derived from improper prior measures are found to have an unBayesian property—one that could not occur if proper prior measures were employed. This paradoxical possibility is shown to have several facets that can be successfully analysed in the framework of a general group structure. The results cast a shadow on the uncritical use of improper prior measures. A separate examination of a particular application of Fraser's structural theory shows that it is intrinsically paradoxical under marginalization.Keywords
This publication has 21 references indexed in Scilit:
- Expectation consistency of inverse probability distributionsBiometrika, 1972
- Structural Distributions Without Exact TransitivityThe Annals of Mathematical Statistics, 1972
- Fiducial Theory and Invariant EstimationThe Annals of Mathematical Statistics, 1966
- Bayesian Estimation in Multivariate AnalysisThe Annals of Mathematical Statistics, 1965
- Introduction to Probability and Statistics from a Bayesian Viewpoint1965
- Invariant Prior DistributionsThe Annals of Mathematical Statistics, 1964
- Bayesian statistical inference for psychological research.Psychological Review, 1963
- On Fiducial InferenceThe Annals of Mathematical Statistics, 1961
- Some Problems Connected with Statistical InferenceThe Annals of Mathematical Statistics, 1958
- A Note on the Generation of Random Normal DeviatesThe Annals of Mathematical Statistics, 1958