Estimation of semi-functional partially linear models with censored functional covariates

In this paper,we study the estimation problem of semi-functional partial linear models under the background of censored function covariates,and extend censored function data to complete function data by using a curve extension algorithm. The algorithm has good accuracy and flexibility§ and avoids the problem that the censored function data is difficult to model. The estimates of unknown parameters and smoothing operators in the model are obtained by least square estimation and functional kernel estimation§respectively. The effectiveness of the proposed algorithm is verified by numerical simulation,and applied to data analysis of Tecator and Primary Biliary Cirrhosis data sets.