A Comparison of Two Methods for Estimating Censored Linear Regression Models

Authors

  • Ersin Yılmaz Mugla Sitki Kocman University, Turkey
  • Dursun Aydın Mugla Sitki Kocman University, Turkey

Keywords:

Least squares, Kaplan-Meier estimator, linear regression, right-censored data, synthetic data, Kaplan-Meier weights

Abstract

This paper presents two basic methods called as weighted least squares (WLS) and synthetic data transformations (SDT). The key idea of the paper is to estimate the parameters of the linear regression model with randomly right-censored data by using these two methods. Recently, the mentioned methods have received considerable attention in the literature. Studies on this subject show that both methods work well for linear regression model with censored data. A particular focus of our paper is to compare the performance of the WLS and SDT methods and to reveal the strong and weak aspects of them. In this context, we made a simulation study and a real data example.

Downloads

Published

2017-08-30

Issue

Section

Articles