Kernel regression for cause-specific hazard models with time-dependent coefficients

TL;DR

This paper introduces a kernel likelihood method for estimating cause-specific hazard models with time-dependent coefficients in competing risk data, demonstrating good performance through simulations and real data application.

Contribution

It develops a novel nonparametric kernel estimator for time-varying coefficients in cause-specific hazard models, filling a gap in existing methods.

Findings

Kernel estimator performs well in finite samples

Method successfully applied to diabetes dialysis data

Provides flexible modeling of time-dependent effects

Abstract

Competing risk data appear widely in modern biomedical research. Cause-specific hazard models are often used to deal with competing risk data in the past two decades. There is no current study on the kernel likelihood method for the cause-specific hazard model with time-varying coefficients. We propose to use the local partial log-likelihood approach for nonparametric time-varying coefficient estimation. Simulation studies demonstrate that our proposed nonparametric kernel estimator has a good performance under assumed finite sample settings. Finally, we apply the proposed method to analyze a diabetes dialysis study with competing death causes.