Wild bootstrap logrank tests with broader power functions for testing\n superiority

2018/08/16 by Marc Ditzhaus, Markus Pauly, Ditzhaus, Marc +1
Mathematics · #Advanced Statistical Methods and Models #FOS: Mathematics #Statistical Methods and Inference #Statistical Methods in Clinical Trials #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.1808.05627

openalex publication_date 2018/08/16 · openalex created_date 2022/08/04 · openalex updated_date 2026/07/28

Abstract

We introduce novel wild bootstrap procedures for testing superiority in\nunpaired two-sample survival data. By combining different classical weighted\nlogrank test we obtain tests with broader power behavior. Right censoring\nwithin the data is allowed and may differ between the groups. The tests are\nshown to be asymptotically exact under the null, consistent for fixed\nalternatives and admissible for a larger set of local alternatives. Beside\nthese asymptotic properties we also illustrate the procedures' strength in\nsimulations for finite sample sizes. The tests are implemented in the novel\nR-package mdir.logrank and its application is demonstrated in an exemplary data\nanalysis.\n

Related

• More powerful logrank permutation tests for two-sample survival data
• A monotonicity property of weighted log-rank tests
• Randomization-based Test for Censored Outcomes: A New Look at the Logrank Test
• Balancing events, not patients, maximizes power of the logrank test: and other insights on unequal randomization in survival trials
• The Anytime-Valid Logrank Test: Error Control Under Continuous Monitoring with Unlimited Horizon