Study of Small Sample Test Data Estimation Base on Bayes Bootstrap Method
Jie Yan · Jisuanji fangzhen · 2010
Bayesian Bootstrap method is one of the methods to evaluate the small sample test data. As has well generality,it is widely used in engineering practice. But as this method does not make use of the prior information,it also has some limitation in practice,so the precision of estimation is limited. This paper puts forward two ways to solve the problem. One is the method which bases on the prior information of probability density function to generate random weighting values,and the other bases on the prior information of distribution function. We also analyze the feasibility of the two methods,and prove the effectiveness of them,which increases the precision of the estimation in the simulation in normal distribution.