Characterization of priors under which Bayesian and frequentist Bartlett corrections are equivalent in the multiparameter case

Journal of Multivariate Analysis

Volumn 38, Issue 2, 1991, Pages 385-393

  Ghosh, J K ^a,b   Mukerjee, Rahul ^c  

^a INDIAN STATISTICAL INSTITUTE   (India)

^b PURDUE UNIVERSITY   (United States)

^c INDIAN INSTITUTE OF MANAGEMENT CALCUTTA   (India)

Author keywords

Bartlett correction; Jeffreys' prior; likelihood ratio test; non informative prior

Indexed keywords

EID: 0001500591     PISSN: 0047259X     EISSN: 10957243     Source Type: Journal    
DOI: 10.1016/0047-259X(91)90052-4     Document Type: Article

References (15)

1. A note on the calculation of Bartlett adjustments
2. On the level error after Bartlett adjustment of the likelihood ratio statistic
3. A decomposition for the likelihood ratio statistic and the Bartlett correction—A Bayesian argument
4. The convergence of Bhattacharya bounds
5. Valid asymptotic expansions for the likelihood ratio statistic and other perturbed chi-square variables
6. Parameter orthogonality and approximate conditional inference (with discussion)
7. Invariant prior distributions
8. Convergence of Bhattacharya bounds-revisited
9. Expansions for posterior probability and integrated Bayes risk
10. Asymptotic expansions associated with posterior distributions
11. Natural exponential families with quadratic variance functions
12. On the variance of estimates
13. Some characterizations based on the Bhattacharya matrix
14. On coverage probability of confidence sets based on a prior distribution
15. On formulae for confidence points based on integrals of weighted likelihoods