The Non-stationary Of Markov Process With Generalized Poisson Marginal Distribution Based On Binomial Thinning

What is known to us all that the problem of elliptic equation is one of the interests of many researchers,and many authors does a lot in this field.A large number of classical results have been obtained with aid of their continu-ous researching and exploration.This paper studies a non-stationary Markov one order time series model based on binomial thinning,whose marginal dis-tribution is the generalized poisson distribution.The simplifies properties and estimation for the classes of time series model that we consider.And also gives the numerical simulation.An application with real date is used to illustrate the model.This paper is organized as follows.The first chapter is the introduction,which introduces the background,main results of this paper.In the second chapter,we introduce the knowledge of model and it's character.In order to better adapt to the over-dispersion data.We change the negative binomial to generalized poisson.Then in the third chapter,the estimation of parameters are carried out.Then,we used the R's optim function to solves the equations that cannot be solved for explicit parameter estimation expressions.The pa-rameters of the model are simulated numerically.In the fourth chapter,Two application with real date is used to illustrate the model.The fifth chapter is a summary.