Nonparametric regression function estimation with surrogate data and validation sampling

摘要: This paper develops estimation approaches for nonparametric regression analysis with surrogate data and validation sampling when response variables are measured errors. Without assuming any error model structure between the true responses variables, a calibration kernel estimate is defined help of data. The proposed estimator proved to be asymptotically normal convergence rate also derived. A simulation study conducted compare estimators standard Nadaraya-Watson observations in set complete observations, respectively. can serve as gold standard, even though it practically unachievable because measurement