定数截尾两参数指数——威布尔分布形状参数的Bayes估计

The Bayesian Estimation of Shape-Parameter of Two-Parameter Exponentiaed-Weibull Distribution Under the Type-Ⅱ Censoring Test

在不同的损失函数下,本文研究了两参数指数—威布尔分布(EWD)形状参数的Bayes估计问题.基于定数截尾试验,当其中一个形状参数α已知时,给出了另一个形状参数θ在三种不同损失函数下的Bayes估计表达式,并求得了可靠度函数的Bayes点估计.最后运用随机模拟方法,将Bayes估计和极大似然估计进行了比较.结果表明,LINEX损失下Bayes估计的精度比极大似然估计高.

In this paper we study the Bayes estimations of shape-parameter of two-parameter EWD under several different loss functions. Based on type-Ⅱ censoring test, when one shapeparameter is fixed, the Bayesian estimations of another shape-parameter under several loss functions, are given. Also we offer the reliability function an expression of Bayes point estimation. Then we compare the Bayes estimations with the MLE by means of Monte Carlo simulation. The results show that the accuracy of the Bayes estimation under LINEX loss is better than that of the MLE.