Inferring median survival differences in general factorial designs via\n permutation tests

2020/06/25 by Marc Ditzhaus, Ditzhaus, Marc, Dennis Dobler +3
Decision Sciences · Mathematics · #FOS: Computer and information sciences #Methodology (stat.ME) #Optimal Experimental Design Methods #Statistical Distribution Estimation and Applications #Statistical Methods in Clinical Trials

paper · pdf · doi:10.48550/arxiv.2006.14316

openalex publication_date 2020/06/25 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

Factorial survival designs with right-censored observations are commonly\ninferred by Cox regression and explained by means of hazard ratios. However, in\ncase of non-proportional hazards, their interpretation can become cumbersome;\nespecially for clinicians. We therefore offer an alternative: median survival\ntimes are used to estimate treatment and interaction effects and null\nhypotheses are formulated in contrasts of their population versions.\nPermutation-based tests and confidence regions are proposed and shown to be\nasymptotically valid. Their type-1 error control and power behavior are\ninvestigated in extensive simulations, showing the new methods' wide\napplicability. The latter is complemented by an illustrative data analysis.\n

Related

• Permutation inference in factorial survival designs with the CASANOVA
• Survival of the Fittest Group: Factorial Analyses of Treatment Effects\n under Independent Right-Censoring
• A Multiple kernel testing procedure for non-proportional hazards in factorial designs
• Comparing restricted mean survival times in small sample clinical trials using pseudo-observations
• Studentized Permutation Method for Comparing Restricted Mean Survival Times with Small Sample from Randomized Trials