Asymptotic Expansion of D-Risks for Hypothesis Testing in Bernoulli Scheme
Currently there is not a single asymptotic expansion of d-risk of any statistical procedure. In this paper we present the asymptotic expansion of d-risk function of the optimal test for Bernoulli scheme. The expansion derivation is based on the Edgeworth series of the sufficient statistic for Bernoulli scheme and therefore it is applicable to any one-parametric exponential model.
KeywordsBayesian analysis d-risk asymptotic expansion Bernoulli distribution
Unable to display preview. Download preview PDF.
- 10.A. A. Zaikin, “Asymptotic expansion of posterior distribution of parameter centered by n-consistent estimate,” Zap. Nauch. Sem. SPb. Otd. Mat. Inst. Steklova POMI 454, 121–150 (2016).Google Scholar
- 11.S. V. Simushkin, “Optimal d-guaranteed procedures of testing two hypotheses,” Available from VINITI No. 5547-81 (1981).Google Scholar
- 16.R. N. Bhattachariya and R. R. Rao, Normal Approximation and Asymptotic Expansions (SIAM, Philadelphia, 1986).Google Scholar