A semi-parametric unified cure model for interval censored data

Article Type

Research Article

Publication Title

Communications in Statistics Simulation and Computation

Abstract

We extend the unified class of Box-Cox transformation cure model (BCTM) to accommodate interval censored data. To capture the heterogeneity in the survival distribution of the uncured subjects, we use a proportional hazards (PH) structure. Furthermore, to avoid any theoretical distribution based parametric assumption we approximate the baseline hazard using piecewise linear functions. To estimate the model parameters, we develop an expectation-maximization (EM) algorithm where a profile likelihood (PL) approach is implemented to estimate the BCTM index parameter. Our simulation study results show that the proposed estimation method performs well in retrieving the true parameter values. A comparison with the simultaneous maximization of all model parameters within the EM framework suggests that the proposed EM-based PL approach performs better. Finally, the proposed model and methods are applied to analyze a real interval censored smoking cessation data.

DOI

10.1080/03610918.2025.2596299

Publication Date

1-1-2025

Link to Full Text

Share

COinS