q-对称熵损失函数下Gamma分布的尺度参数的估计  

Estimation of Scale Parameters of the Gamma Distribution under the Q-symmetric Entropy Loss Function

作　　者：杜宇静[1] 孙晓祥[1] 万喜昌[1] 

机构地区：[1]吉林农业科技学院基础部,吉林132101

出　　处：《工程数学学报》2008年第3期500-504,共5页Chinese Journal of Engineering Mathematics

基　　金：国家自然科学基金(10571073).

摘　　要：本文在对称熵损失函数的基础上定义了q-对称熵损失函数,并用参数估计的方法研究了在q-对称熵损失函数下Gamma分布的尺度参数的最小风险同变估计(MRE)、贝叶斯(Bayes)估计、最小最大(Mininax)估计等。我们还对这些估计量的可容许性和不可容许性进行了讨论,最后分别对指数分布和Gamma分布在两种损失函数下的估计结果进行了数值比较。In this paper,we define the q-symmetric entropy loss function on the basis of the symmetric entropy loss function. A parametric estimation method for the analysis of the scale parameter of the Gamma distribution is presented, we also deal with the minimum risk equivalent estimation, the Bayesian estimation and the Minimax estimation under the proposed loss function. Furthermore, the admissibility and inadmissibility of these estimators are discussed. Numerical comparison is carried out about the estimates under the exponential these loss distribution and the Gamma loss distribution, respectively.

关 键 词：贝叶斯估计 同变估计 最小最大估计 尺度参数 可容许性 Q-对称熵损失函数 

分 类 号：O212.5[理学—概率论与数理统计]