Bayes Analysis Of Bivariate Inverse Weibull Distribution

The inverse Weibull distribution is widely used in survival analysis.It also has good re-search prospects.This paper mainly discusses the parameter estimation of bivariate inverse Weibull distribution.In this paper,we first introduce the definition of bivariate inverse Weibull distribution,dis-tribution function,and x₁<x₂,x₁>x₂,x₁=x₂=x probability density function.Suppose there is an observation sample from bivariate inverse Weibull whose capacity is n,the introduction of potential variables y and z,construct the like function of bivariate inverse Weibull dis-tribution.Then realize the maximum like estimation of parameters on the basis of EM algorithm,and using the Gibbs sampling algorithm,the Bayes estimation of unknown parame-ters of bivariate inverse Weibull distribution is realized.According to the simulation results,we can find that with the increase of sample size,the maximum like estimation and Bayes estimation are similar,and the results of the two parameter estimation methods are good.But in small samples,the Bayes estimation based on Gibbs sampling algorithm has better effect.