Estimators Of The Error Variance In Logistic Regression Models

Generalize linear model is a kind of widely used model,which is a generalization of general linear model,especially in discrete data.The empirical likelihood method is one of the most popular statistical inference methods in recent years.As it does not require estimation variance,it has attracted the interest of many statisticians,who applied it to various statistical model and various domains.Therefore,it is of great significance for us to study the estimation of error variance of Logistic regression model by using the empirical likelihood method.In this paper,empirical likelihood method is used to study the estimation of error variance of Logistic regression model.In the second chapter,based on random sample,we obtain the asymptotic normality of error variance of Logistic regression model under two different estimations.And calculated Logistic model variance estimator variance of the theoretical value,by combining it with traditional estimation error variance estimators of variance under the theoretical value comparison we found that using the empirical likelihood method to get the random design case Logistic model variance is smaller than the traditional estimation error variances ?n2 variance;At the same time,we also use data simulation to verify this conclusion,and get the same conclusion as our theory,that is to say,the empirical likelihood estimation method improves the estimation accuracy of Logistic model error variance under a certain situation.