Semiparametric Model Based On Empirical Likelihood Statistical Diagnosis

In this paper, the empirical likelihood method to study the semi-parametric statistical model to diagnose problems. First through the semi-parametric model of kernel estimation to estimate the unknown function, then the parameter estimation equation based on the experience of application of likelihood methods to estimate, and discussed the asymptotic properties of estimates, given the approximate parameter confidence region. Experience based on the deletion model proposed experience like natural Cook distance, likelihood distance and standardized residuals,then find outliers and strong impact point. Finally, a simulation example and a real problem in the data to verify the feasibility and effectiveness.