摘要
为了充分利用参数的先验信息,研究了一元P-范分布尺度参数的经验贝叶斯估计问题。利用先验选择的矩方法确定先验分布的超参数,在平方损失函数下获得了一元P-范分布尺度参数的经验贝叶斯估计,并进一步证明了其渐近最优性,同时讨论了估计的容许性。
In order to make full use of the prior information of the parameters,the parametric empirical Bayes estimation of scale parameter for one variable P-norm distribution is investigated.Moment method of choosing prior distribution is used to determine the parameter of prior distribution.Empirical Bayes estimation of scale parameter for one variable P-norm distribution under quadratic loss function and its asymptotically optimality and admissibility are given.
引文
[1]童光荣,李思维.偏态P范分布刻画股指收益率的实证检验[J].统计与决策,2015(8):167-169.
[2] SUN Haiyan.A more GENERAL form of error distribution[J].Geomatics and Information Science of Wuhan University,1993,18(3):49-56.
[3]潘雄,罗静,刘衍宏,等.基于矩估计法的P范分布参数估计[J].武汉大学学报(信息科学版),2017,42(4):563-568.
[4] WANG Chao,CHEN Jiaqing,LIU Cihua,et al.Empirical Bayes estimation of the scale parameter for a generalized exponential distribution under entropy loss[J].Mathematica Applicata,2016,16(2):35-40.
[5] TAN Zhiqiang.Steinized empirical Bayes estimation for heteroscedastic data[J].Statistica Sinica,2016,26(3):55-61.
[6] KIM J E,NEMBHARD D A.Parametric empirical Bayes estimation of individual time-pressure reactivity[J].International Journal of Production Research,2017(3):1-12.
[7]陈家清,王玉,陈志强.熵损失函数下Burr(α)X分布参数的贝叶斯估计[J].统计与决策,2017(20):5-9.
[8]陈家清,胡锦霞,刘次华.基于Stein损失污染数据情形下刻度参数的经验贝叶斯估计[J].应用数学,2017(3):562-569.
[9]温利民,张美,程子红,等.帕累托索赔分布中风险参数的经验贝叶斯估计[J].应用概率统计,2015,31(3):225-237.
[10]尤游,周玲.LINEX损失下双指数分布位置参数的经验Bayes估计[J].应用概率统计,2015,31(3):254-266.
[11]茆诗松,王静龙,濮晓龙.高等数理统计[M].2版,北京:高等教育出版社,2006:115-116.