Nonparametric regression estimates based on imputation techniques for right-censored data

Abstract

Censored data is a kind of data type where the exact value of a response variable is not completely known. Therefore, this case is a problem that should be solved in order to obtain an accurate and efficient data analysis. Recently, imputation methods have been used in order to overcome censored data problems, especially in medical research and microarray data sets. In this study, we compared two imputation methods, k-nearest neighbors (kNN) and a prediction model (PM), for the evaluation of right-censored data. In order to see the effects of the imputation methods on the nonparametric regression estimates, the imputed right-censored data modelled by the penalized splines for two methods. We also supported the study with a Monte Carlo simulation experiment and a real data study. © Springer Nature Switzerland AG 2020.