Diagnostic And Empirical Likelihood For Semi-parametric Nonlinear Model With Missing Data

Missing response data often arise in various experimental settings,including market research surveys,medical studies,opinion polls and socioeconomic investigations.Statistical analysis with missing data is a very difficult task since in most cases the missing data themselves contain little or no information about the missing data mechanism.lt has been widely used to deal with missing data by means of loan and complement under the MAR assumption.This paper discusses the statistical diagnosis of semi parametric nonlinear model with missing data in the case of the response or covariate.First,the randomly missing data is supplemented and the complete sample data set is obtained by the method of the modified borrowing and supplementing.Then,through kernel estimation to estimate the unknown function,and the empirical likelihood method is used to estimate the parameter estimation of the modeI·According to the data deletion model,the relationship of ? and ?(t)is derived.Some diagnostic statistics are proposed in order to find out the problem points in the sample data,that is outlier point or strong influence point.Finally,a real data is given to illustrate the validity of statistical diagnostic measures.