Semiparametric Estimation for Cure Survival Model with Left-Truncated and Right-Censored Data and Covariate Measurement Error

TL;DR

This paper develops a semiparametric approach for cure survival models that addresses left-truncated data and measurement error in covariates, providing theoretical guarantees.

Contribution

It introduces a flexible method to analyze complex survival data with left-truncation and measurement error, along with theoretical validation.

Findings

Proposed a new semiparametric estimation method

Corrected for measurement error in covariates

Established theoretical properties of the estimators

Abstract

In this paper, we mainly discuss the cure model with survival data. Different from the usual survival data with right-censoring, we incorporate the features of left-truncation and measurement error in covariates. Generally speaking, left-truncation causes a biased sample in survival analysis; measurement error in covariates may incur a tremendous bias if we do not deal with it properly. To deal with these challenges, we propose a flexible way to analyze left-truncated survival data and correct measurement error in covariates. The theoretical results are also established in this paper.