为了度量E-Bayes估计的误差,该文基于E-Bayes估计的定义,引入了E-Bayes估计的E-MSE(expected mean square error)的定义.对Poisson分布的参数,在不同损失函数(包括:平方损失,K-损失,加权平方损失)下分别给出了E-Bayes估计及其E-MSE的表...为了度量E-Bayes估计的误差,该文基于E-Bayes估计的定义,引入了E-Bayes估计的E-MSE(expected mean square error)的定义.对Poisson分布的参数,在不同损失函数(包括:平方损失,K-损失,加权平方损失)下分别给出了E-Bayes估计及其E-MSE的表达式.用MonteCarlo方法进行模拟比较提出的估计方法的性能,分析了一个真实数据集并进行了比较,所得结果比较基于E-MSE,结果表明该文提出的方法可行且便于应用.展开更多
This paper introduces a new method, E-Bayesian estimation method, to estimate the reliability in zero-failure data. The definition of E-Bayesian estimation of the reliability is given. Based on the definition,the form...This paper introduces a new method, E-Bayesian estimation method, to estimate the reliability in zero-failure data. The definition of E-Bayesian estimation of the reliability is given. Based on the definition,the formulas of E-Bayesian estimation and hierarchical Bayesian estimation of the reliability are provided, and property of the E-Bayesian estimation, i.e. relation between E-Bayesian estimation and hierarchical Bayesian estimation, is discussed. Calculations performed on practical problems show that the proposed new method is feasible and easy to operate.展开更多
文摘为了度量E-Bayes估计的误差,该文基于E-Bayes估计的定义,引入了E-Bayes估计的E-MSE(expected mean square error)的定义.对Poisson分布的参数,在不同损失函数(包括:平方损失,K-损失,加权平方损失)下分别给出了E-Bayes估计及其E-MSE的表达式.用MonteCarlo方法进行模拟比较提出的估计方法的性能,分析了一个真实数据集并进行了比较,所得结果比较基于E-MSE,结果表明该文提出的方法可行且便于应用.
基金the Ningbo University of Technology Science Foundation and Ningbo Natural Science Foundation(No.2013A610108)
文摘This paper introduces a new method, E-Bayesian estimation method, to estimate the reliability in zero-failure data. The definition of E-Bayesian estimation of the reliability is given. Based on the definition,the formulas of E-Bayesian estimation and hierarchical Bayesian estimation of the reliability are provided, and property of the E-Bayesian estimation, i.e. relation between E-Bayesian estimation and hierarchical Bayesian estimation, is discussed. Calculations performed on practical problems show that the proposed new method is feasible and easy to operate.
