On the asymptotic defect of some Bayesian criteria*
- 26 Downloads
This paper focuses on the problem of testing a simple hypothesis about a one-dimensional parameter against one-sided alternatives with independent identically distributed random variables. A formula for extreme deviation of the power function from the envelope power function is obtained for asymptotically efficient Bayesian criteria based on a Bayesian likelihood ratio. This formula makes it possible to find the asymptotic deficiency in terms of the Hodges–Lehmann deficiency. The method used here makes it possible to relax necessary regularity conditions.
KeywordsAsymptotic Expansion Regularity Condition Nuisance Parameter Extreme Deviation Simple Hypothesis
Unable to display preview. Download preview PDF.
- 7.J. Pfanzagl, “Asymptotic expansions in parametric statistical theory,” in: Development in Statistics, Vol.3, Academic Press, New York (1980), 1–97.Google Scholar
- 8.D. M. Chibisov, “Asymptotic expansions and deficiencies of tests,” in: Proc. Intern. Congress of Mathematicians, Vol. 2, Warszawa (1983), 1063–1079.Google Scholar
- 10.J. Pfanzagl, “On asymptotically complete classes,” Statistical Inference and Related Topics, Vol.7, Academic Press, New York (1975), 1–43.Google Scholar
- 11.M. G. Kendall and A. Stuart, Distribution theory, Griffin and Co. (2000).Google Scholar
- 12.D. M. Chibisov, “Asymptotic expansions for Neyman’s C(α) tests,” in: Proc. 2nd Japan-USSR Sypmos. Prob. Theory, Lecture Notes in Mathematics, Vol.330, Springer (1973), 16–45.Google Scholar