Empirical Likelihood Estimators Of The Error Variance In Nonlinear Regression Models

In this paper,the estimation problems of error variances for nonlinear regression model and Probit regression model are studied by using empirical likelihood method respectively.Under certain assumed conditions,the asymptotic normality of the estimator is proved,and it is conclud-ed that the asymptotic variance of the estimator is smaller than that of the traditional estimator.The main conclusions in this paper are verified by numerical simulation,and the efficacies of the estimator proposed in this paper are compared with that of the traditional estimator.The main contents of this paper are as follows:The first chapter is the introduction,which briefly introduces the research situation of non-linear regression model,Probit regression model,empirical likelihood method,and the research content and innovation of this paper.In chapter 2,the error variance of nonlinear regression model is estimated by empirical like-lihood method,and compared with the asymptotic variance of estimators obtained by traditional methods,and the necessary assumptions,main conclusions,simulation results,lemma and the proof of main conclusions are given.In chapter 3,the error variance of Probit regression model is estimated by empirical likeli-hood method,and compared with the asymptotic variance of the estimator obtained by traditional method,and the necessary assumptions,main conclusions,simulation results,lemma and the proof of main conclusions are given.