CAJ | 학술논문

针对下记录值样本，研究了双参数指数威布尔模型的参数估计问题。首先给出了参数的极大似然估计，随后在参数的共轭先验分布为伽玛先验分布、损失函数为平方误差损失和Linex损失函数情形下，应用Lindley ，近似法导出了双参数指数威布尔模型参数的Bayes估计，最后利用R软件进行数值模拟，并对两种参数估计进行比较，发现在合适的先验分布条件下Bayes估计值更加接近参数真实值。因此，在适当的先验分布下Bayes估计较传统的极大似然估计好。
침대하기록치양본，연구료쌍삼수지수위포이모형적삼수고계문제。수선급출료삼수적겁대사연고계，수후재삼수적공액선험분포위가마선험분포、손실함수위평방오차손실화Linex손실함수정형하，응용Lindley ，근사법도출료쌍삼수지수위포이모형삼수적Bayes고계，최후이용R연건진행수치모의，병대량충삼수고계진행비교，발현재합괄적선험분포조건하Bayes고계치경가접근삼수진실치。인차，재괄당적선험분포하Bayes고계교전통적겁대사연고계호。
According to the lower record values of samples ,this paper studies the parameters of two‐parameter exponential Weibull model estimation problem .First ,we give the maximum likelihood estimates of the parameters ,then the parameters of the conjugate prior distribution for the Gama prior distribution .Loss function is the squared error loss and Linex loss function under the conditions of ap‐plication of Lindley approximation ,is derived by estimating parameter Bayes of double parameter ex‐ponential Weibull model .Finally we carry on the numerical simulation using R software .Two kinds of parameters on the proposed estimators are compared ,and we find in the prior distribution of suitable conditions ,Bayes estimates are closer to the true parameters .Therefore the maximum likelihood prop‐er prior distribution estimation is better than the traditional Bayes estimation .