完全隨機(jī)缺失機(jī)制下伽瑪分布中形狀參數(shù)的經(jīng)驗貝葉斯推斷

[1]Johns M V Jr,Van Ryzin J.Convergence rates in empirial bayes two-action Problems 1:Discrete Case[J]. Ann.Math.Statist.,1971,42:1521-1539.

[2]Johns M V Jr,Van Ryzin J.Convergence rates in empirial bayes two-action Problems 2:Continous Case[J]. Ann.Math.Statist.,1972,43:934-937.

[3]Van Hou welingen J C.Monotone empirical Bayes test for the continuous one-parameter exponential family[J]. Ann.Statist.,1976,4:981-989.

[4]Liang Tachen.On optimal convengence rate of empirical Bayes tests[J].Statists and Probability Letters, 2004,68:189-198.

[5]Shang S G,Li J.One Empirical Bayes procedures for selecting good populations in a positive exponential family[J].Journal of Statistical Planning and Inference,2005,129:3-18.

[6]Prakasa Rao B L S.Nonparametric Function Estimation[M].New York:Academic Press,1983.

2000MSC:62C12

Empirical Bayes inference for scale parameter of Gamma distribution with MCAR

LI Nai-yi1,LIU Hua-xiang1,CHEN Ru-yun1,DU Jun2
(1.School of Science,Guangdong Ocean,Zhanjiang524088,China;
2.College of Economics and Management,Guangdong Ocean,Zhanjiang524088,China)

In MCAR,the Empirical Bayes two-sided test rules for parameter of the gamma distribution family are constructed,and the asymptotically optimal property is obtained.It is shown that the convergence rates of the proposed EB test rules can arbitrarily close to O(n?21)under suitable conditions.

MCAR,Empirical Bayes test,asympototic optimality,convergence rates