Wild Bootstrap based Confidence Bands for Multiplicative Hazards Models

TL;DR

This paper introduces wild bootstrap methods to construct reliable simultaneous confidence bands for cumulative hazard functions in multi-state Cox models, accommodating complex data scenarios like censoring and truncation.

Contribution

The paper presents novel resampling techniques for confidence bands in multi-state Cox models, extending existing methods to handle time-dependent covariates and various data censoring mechanisms.

Findings

Methods show good finite sample performance in simulations

Approach successfully applied to empirical data

Provides asymptotically valid confidence bands

Abstract

We propose new resampling-based approaches to construct asymptotically valid time simultaneous confidence bands for cumulative hazard functions in multi-state Cox models. In particular, we exemplify the methodology in detail for the simple Cox model with time dependent covariates, where the data may be subject to independent right-censoring or left-truncation. In extensive simulations we investigate their finite sample behaviour. Finally, the methods are utilized to analyze an empirical example.