Empirical Likelihood Inference Of Error Variance In Spatial Autoregressive Models

Spatial autoregressive model is one of the most basic models in spatial econometrics.It has been developing in theoretical research and application field,and has gradually become the mainstream of econometrics research.Since Owen(1988)proposed the empirical likelihood method,it has been widely used because of its outstanding inferential advantage.However,the current research on empirical likelihood estimation of model error variance has not been involved in the field of spatial econometrics.Therefore,it is significant to infer the error variance of spatial autoregressive model by empirical likelihood method.In this paper,under the background of empirical likelihood method,we study the estimator of error variance of spatial autoregressive model.Under given conditions,we prove the asymptotic normality of the estimator,and then compare the asymptotic variance of the traditional estimator to obtain the asymptotic variance of the estimator,It is proved that under certain conditions the asymptotic variance of the estimation obtained by the empirical likelihood method is smaller than that of the traditional estimation.Finally,through the stochastic simulation compared the empirical likelihood method in this paper and the advantages and disadvantages of the traditional method,the simulation results verify the main conclusion of this article.The organizational structure of this paper is as follows:The first chapter is the introduction,the research status of spatial autoregressive model,the background and progress of empirical likelihood method,the main content and innovation of this paper.The second chapter is the main part of the paper.The empirical likelihood method is used to estimate the error variance of the spatial autoregressive model.The results are compared with the asymptotical variance estimated by the traditional method.The third chapter is the main conclusion and prospect.