We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ME

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Methodology

Title: BONuS: Multiple multivariate testing with a data-adaptivetest statistic

Abstract: We propose a new adaptive empirical Bayes framework, the Bag-Of-Null-Statistics (BONuS) procedure, for multiple testing where each hypothesis testing problem is itself multivariate or nonparametric. BONuS is an adaptive and interactive knockoff-type method that helps improve the testing power while controlling the false discovery rate (FDR), and is closely connected to the "counting knockoffs" procedure analyzed in Weinstein et al. (2017). Contrary to procedures that start with a $p$-value for each hypothesis, our method analyzes the entire data set to adaptively estimate an optimal $p$-value transform based on an empirical Bayes model. Despite the extra adaptivity, our method controls FDR in finite samples even if the empirical Bayes model is incorrect or the estimation is poor. An extension, the Double BONuS procedure, validates the empirical Bayes model to guard against power loss due to model misspecification.
Subjects: Methodology (stat.ME); Statistics Theory (math.ST)
Cite as: arXiv:2106.15743 [stat.ME]
  (or arXiv:2106.15743v2 [stat.ME] for this version)

Submission history

From: Chiao-Yu Yang [view email]
[v1] Tue, 29 Jun 2021 22:08:26 GMT (1273kb,D)
[v2] Thu, 1 Jul 2021 18:28:07 GMT (1257kb,D)

Link back to: arXiv, form interface, contact.