基于插值法的Bayes Bootstrap方法的改进  

Improvement of Bayes Bootstrap Method Based on Interpolation Method

作　　者：孙慧玲[1] 胡伟文[1] 刘海涛[1] Sun Huiling Hu Weiwen Liu Haitao(College of Science, Naval University of Engineering, Wuhan 430033, Chin)

机构地区：[1]海军工程大学理学院,武汉430033

出　　处：《统计与决策》2017年第9期74-77,共4页Statistics & Decision

基　　金：国家自然科学基金资助项目(61074191);海军工程大学自主立项青年项目(425517K107)

摘　　要：小样本实验数据较小,工程上常用Bayes Bootstrap方法来估计小样本的参数问题,文章研究了置信度一定、样本量不同的情况下,用插值法自助扩充了小样本数据的样本量,对Bayes Bootstrap方法提出改进。最后,运用蒙特卡罗方法对具体算例建模仿真,验证改进方法的优越性。Small-sized samples offer small experimental data and in engineering Bayes Bootstrap method is often used to estimate parameters of small-sized samples. This article tries an improvement to the Bayes Bootstrap method by using interpolation method to extend the sample capacity of small-sized sample data in the case of the fixed confidence coefficient but with different sample size. Finally, the Monte Carlo （MC） simulation is adopted to model and simulate the specific numerical examples, through which the effectiveness and practicability of the improved-Bootstrap method is verified.

关 键 词：小样本 插值法 BAYES BOOTSTRAP方法 M-C仿真 

分 类 号：O241.3[理学—计算数学]