摘要
针对正态分布下线性无偏估计在PITMAN准则下的比较问题,提出了一种较简单的比较方法,解决了无法在PITMAN准则下比较线性无偏估计的问题。将线性无偏估计在PITMAN准则下的优良性转换为比较线性无偏估计在均方误差下的优良性。该方法也可用于常见的线性回归模型中几类线性无偏估计的比较。
In this paper,we propose a simple method to compare the linear unbiased estimators under the Pitman's measure of closeness for the normal distribution. This method solved the literature that the linear unbiased estimators can not be compared under the Pitman's measure of closeness. This method converts the superiority of linear unbiased estimation under PITMAN criterion into the superiority of linear unbiased estimation under mean square error. This method can also be used to compare several types of linear unbiased estimators in common linear regression models.
引文
[1]RAO C R.Some comments on the minimum mean square error as a criterion of estimation[J].Statistics and related topics,1980:123-143.
[2]PITMAN E J G.The closest estimates of statistical parameters[J].Proc Cambridge Philosop Soc,1937,33:212-222.
[3]WANG S G,YANG H.Linear estimators umder pitman nearness criterion[J].Chinese Science Bulletin,1994,40(23):1944-1948.
[4]YANG H,LI W X,XU J W.Comparison of two estimators of parameters under Pitman nearness criterion[J].Communications in Statistics-Theory and Methods,2010,39(17):3081-3094.
[5]JOZANI M J.A note on Pitman’s measure of closeness with balanced loss function[J].Statistics,2014,48:552-557.
[6]WU J B.The small sample properties of the restricted principal component regression estimator in linear regression model[J].Commun Statist Theor Meth,2017,46(4):1661-1667.
[7]MASON R L,KEATING J P,SEN P K,BLAYLOCK NW.Comparison of linear estimators using pitman measure of closeness[J].J Amer Statist Assoc,1990,85:579-581.
[8]YAN L Q.A Note on Comparison of(Non)Linear Estimators Using Pitman’s Measure of Closeness[J].Commun Statist Theor Meth,2007,36:2687-2694.
[9]FOUNTAIN R L,KEATING J P.The pitman comparison of unbiased linear estimators[J].Statist Probab Lett,1994,19:131-136.
[10]WANG S G,SHOW S C.Advanced Linear models[M].New York:Marcell Dekker Inc,1993.