Based On The Bayesian Survival Test Censored Regression Model And Its Application

With the development of science and technology, the people attach more and more importance to the reliability of production. Beacuse the life of the production is a stochastic phenomenon, it comes to a statistical inference problem that we try to determine the reliability index of the production. In order to make clear the life distribution of the production measured and solve the indexes of reliability and study the invalid mechanism of the production to give some pieces of advice about improving the reliability of the production. we often have to make the life experiment. The life experiment is separated into two kinds: complete life experiment and"cutting tail"experiment. but the later one is used widely.When we deal with the practical problem by the statistical methods, we often encounter the trouble about missing data. For example, during the experiment on the life of production. Some data of experiment are missed or unobserved owing to the equipment of experiment and observation means or other difficulties. So we get"missing data". It is not worth and unable to make the experiment again as the result of the parts of missing data. It is a special and more difficult question how to analyse statistically about the existing data after"missing data"in survival analysis.Under the condition of censoring data, the trust the people infer the index of reliability of production if only according to the information of the censored data if influenced. This paper discusses the analysis of the survival of three models : Extreme regression model, Weibull regression model and the sub-index sharing heterogeneous model. Fully utilize the existing information in the circumstances, given the Bayesian estimation parameters. But it was difficult to calculate, because censored data in the form of samples of the likelihood function is very complex. To calculate the simple, using Gibbs sampling come to the posterior distribution. WinBUGS use simulation software for data analysis. WinBUGS three procedures were designed to model and model to a non-cycle analysis, it was found that a posteriori estimate parameters related statistics and related statistical plots. Survival analysis shows three models in the effectiveness and reliability.