Hierarchical Bayesian method on the basis of maximum posterior risk;

This paper gives a Bayesian plan of qualification test for production in binomial case by the reliability growth model, and the maximum posterior risk criterion for the Bayesian qualification test is proposed.

A plan of reliability qualification test for protecting the benefits of user is formulated by using Bayesian method in the view of maximum posterior risk.

Under the type-II censoring test,making use of decision theory,fully utilizing the prior information of product,considering experimental expense and times,and combining the loss function,the decision-making scheme of minimum posterior risk for Pareto distribution is worked out.

This paper puts forward minimum posterior risk to choose the superior failure distribution of products based on of Bayesian decision,and then resolves the problem about how to draw up the optimal distribution model from Multi-model,especially,the method is uniform,the standard is consistent and the operation is simple compared with the traditional hypothesis test.

Classical risk,relative risk and after-test risk are usually used to measure test project in the course of making test project of reliability based on Bayes.

Let G_0 and G_1 be two populations having logarithmic normal distribution with parameters(μ_0,σ) and(μ_1,σ) for the life data in an experiment,the Bayesian methods of the classification problem are given in this paper according to the rule of minimum posterior risk,and a new rule is provided for the life classification.

μ=μ_1(μ_1>μ_0>0) are given in this paper according to the rule of minimum posterior risk,and a new rule is provided for the life classification.

This paper describes a novel speaker adaptation framework that combines the maximum a posteriori (MAP) estimation and wighted neighbor regression (WNR) methods.

To solve the problem of poor resolution in passive millimeter wave(PMMW) imaging,we present an improved maximum a posteriori(MAP) super-resolution algorithm.