Applications of empirical likelihood to nonresponse problem and changepoint detection

In this dissertation, we propose an empirical likelihood based method to solve the nonresponse problem and changepoint detection problem. Both methods avoid potential model misspecification problems from which existing parametric methods may suffer. Moreover, the proposed imputation method can correct the bias of the estimator for the parameter of interest for distributions withunder- or over-dispersion problem. In addition, the empirical likelihood changepoint detection method is able to detect the change in parameters other than the population mean.