Bayesian Estimation and MCMC Sampling for the Mortality Probability Model of Population

Abstract

In this paper we consider the mortality probability model of population throughout the whole lifespan based on the mortality rate in the Weibull distribution, that is, the exponential constant of the mortality rate is transformed into a variable function. In the paper we provide the variable function with alternative methods. However the precision of linear function is less than that of nonlinear function with the same number of parameters. The paper proposes a nonlinear variable function as the exponential of mortality rate. Bayesian estimation provides a feasible treatment of the complicated model resorting to MCMC algorithms. Finally we carry out a research into the mortality probability model of Rattus norvegicus population. MC error precision of parameters reaches to 10-4 . The results show us Bayesian estimation is an effective method using MCMC sampling the mortality probability model of population.