负相伴样本情形连续型单参数指数族参数的经验Bayes检验问题

The empirical Bayes test problem for continuous one-parameter exponential family in the case of negatively associated samples

在加权“线性损失”下讨论了负相伴样本情形连续型单参数指数族参数的经验Bayes(EB)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,并获得了它的渐近最优(a.o.)性,在适当的条件下证明了所提出的经验Bayes检验函数的收敛速度可任意接近O(-n 1/2).

By using the kernel-type density estimation in the case of identically distributed and negatively associated samples, the empirical Bayes test rules for the continuous one-parameter exponential family are constructed under weighed linear loss function, 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^-1/2） under suitable conditions.