Optimal weighting for false discovery rate control
By Étienne Roquain, M. V. D. Wiel. 2008
65 citations recorded by the source corpus.
How to weigh the Benjamini-Hochberg procedure? In the con- text of multiple hypothesis testing, we propose a new step-wise procedure that controls the false discovery rate (FDR) and we prove it to be more powerful than any weighted Benjamini-Hochberg procedure. Both finite- sample and asymptotic results are presented. Moreover, we illustrate good performance of our procedure in simulations and a genomics application. This work is particularly useful in the case of heterogeneous p-value distri- butions.
Subfields: Management Science and Operations Research, Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/3082ac002a2bc0eacc462a7d3d5603ccd0560e40
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Management Science and Operations Research
- A. Chakrabarti
- Jayanta Kumar Ghosh
- Florian Frommlet
- Malgorzata Bogdan
- Wolfgang Huber
- Fanny Villers
- Étienne Roquain
- Nikolaos Ignatiadis
In Statistics and Probability
- A. Chakrabarti
- Fanny Villers
- Wolfgang Huber
- Nikolaos Ignatiadis
- Jayanta Kumar Ghosh
- Malgorzata Bogdan
- Florian Frommlet
- Étienne Roquain
Citations in the corpus
- The Bayes oracle and asymptotic optimality of multiple testing procedures under sparsity (2010, methodology): “tiple testing procedures controlling FDR. In recent years many new multiple testing procedures, which have some optimality properties in the classical sense, have been proposed (e.g., [11, 29, 32] or [33]). In [20] an asymptotic analysis is performed and new step-up and step-up-down procedures, whic”
- Covariate powered cross‐weighted multiple testing (2017, background): “er statistics of Q1,...,Qm and let Q(0) := 0 3 Let k∗ = max n k | Q(k) ≤ kα m , k ≥ 0 o 4 Reject all hypotheses with Qi ≤ Q(k∗) the literature to date only considers the case of deterministic weights [8,23,40]. These results are very valuable, since weighted multiple testing procedures have been sho”
- Exact calculations for false discovery proportion with application to least favorable configurations (2010, background): “, [4, 26, 27, 30]), this puts forward a connection between the FDR expression and the rejection number distribution in the step-up case, under independence of the p-values.”