本文研究了NA样本情形下.伽玛分布族形状参数的经验Bayes(EB)双边检验问题.利用概率密度函数的核估计,构造了参数的经验Bayes检验函数.并在适当的条件下,证明了所提出的经验Bayes检验函数的渐近取优(α.o.)性.获得了其收敛速度.
In this paper.the empirical Bayes(EB) two-sided test problem of shape parameter for Gamma distribution families in the case of identically distributed and negatively associated (NA) samples is investigated. By using the kernel-type density estimation, the empirical Bayes two-sided test rules are constructed.the asymptotically optimal property and convergence rates for the proposed EB testrules are obtained under suitable conditions.