Saved in:
Bibliographic Details
Main Authors: Zhuang, Zian, Su, Wen, Kawaguchi, Eric, Li, Gang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.15040
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911066766704640
author Zhuang, Zian
Su, Wen
Kawaguchi, Eric
Li, Gang
author_facet Zhuang, Zian
Su, Wen
Kawaguchi, Eric
Li, Gang
contents Evaluating and validating the performance of prediction models is a fundamental task in statistics, machine learning, and their diverse applications. However, developing robust performance metrics for competing risks time-to-event data poses unique challenges. We first highlight how certain conventional predictive performance metrics, such as the C-index, Brier score, and time-dependent AUC, can yield undesirable results when comparing predictive performance between different prediction models. To address this research gap, we introduce a novel time-dependent pseudo $R^2$ measure to evaluate the predictive performance of a predictive cumulative incidence function over a restricted time domain under right-censored competing risks time-to-event data. Specifically, we first propose a population-level time-dependent pseudo $R^2$ measures for the competing risk event of interest and then define their corresponding sample versions based on right-censored competing risks time-to-event data. We investigate the asymptotic properties of the proposed measure and demonstrate its advantages over conventional metrics through comprehensive simulation studies and real data applications.
format Preprint
id arxiv_https___arxiv_org_abs_2507_15040
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Time-Dependent Pseudo $\boldsymbol{R^2}$ for Assessing Predictive Performance in Competing Risks Data
Zhuang, Zian
Su, Wen
Kawaguchi, Eric
Li, Gang
Methodology
Applications
Evaluating and validating the performance of prediction models is a fundamental task in statistics, machine learning, and their diverse applications. However, developing robust performance metrics for competing risks time-to-event data poses unique challenges. We first highlight how certain conventional predictive performance metrics, such as the C-index, Brier score, and time-dependent AUC, can yield undesirable results when comparing predictive performance between different prediction models. To address this research gap, we introduce a novel time-dependent pseudo $R^2$ measure to evaluate the predictive performance of a predictive cumulative incidence function over a restricted time domain under right-censored competing risks time-to-event data. Specifically, we first propose a population-level time-dependent pseudo $R^2$ measures for the competing risk event of interest and then define their corresponding sample versions based on right-censored competing risks time-to-event data. We investigate the asymptotic properties of the proposed measure and demonstrate its advantages over conventional metrics through comprehensive simulation studies and real data applications.
title Time-Dependent Pseudo $\boldsymbol{R^2}$ for Assessing Predictive Performance in Competing Risks Data
topic Methodology
Applications
url https://arxiv.org/abs/2507.15040