一类低先验信息度的先验分布选择研究  

Study of prior distribution selection with low prior information

作　　者：王维[1] WANG Wei(Aliyun school of Big Data,Changzhou University,Suzhou 213164,China)

机构地区：[1]常州大学阿里云大数据学院,江苏常州213164

出　　处：《陕西理工大学学报（自然科学版）》2021年第3期68-73,共6页Journal of Shaanxi University of Technology:Natural Science Edition

基　　金：常州大学科技计划项目(ZMF20020242)。

摘　　要：以一类低先验信息下参数Bayes估计的先验分布选择为研究对象,采用控制变量法分别研究了样本量、先验分布的类型与方差对估计结果的影响。研究结果表明:对于该类问题,样本量充分大时,先验分布的选择对结果影响不大;样本量较小时,先验分布的类型对结果几乎没有影响;先验分布方差相差越多,结果差异越大。需要指出的是,由于条件所限,无法对所有符合条件的先验分布一一验证,因而尚不能断言所得到的结果具有普适性。The prior distribution selection of parameter Bayes estimation with low prior information is taken as the research object,and the influence of sample size,type of prior distribution and variance on estimation results is studied by using control variable method.The results show that the selection of prior distribution has little effect on the results when the sample size is large enough.When the sample size is small,the type of prior distribution almost has no effect on the results.It should be pointed out that due to the limitation of conditions,this paper can’t verify all the prior distributions that meet the conditions one by one,so we can’t claim that the results obtained are universal.

关 键 词：低先验信息 先验分布 BAYES估计 

分 类 号：O212[理学—概率论与数理统计]