Oracle and Adaptive Compound Decision Rules for False Discovery Rate Control

By Wenguang Sun, Tony Cai. 2007

355 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Molecular Biology, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/fb825a04d32511d4d26372dab63b458e65245010

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

In Molecular Biology

In Statistics and Probability

Citations in the corpus

Multiple testing with the structure‐adaptive Benjamini–Hochberg algorithm (2016, background): “Zhao and Fung [30] allow for this asymmetry by giving distinct weights to p-values associated to positive versus negative statistics, speciﬁcally, by estimating π +0 and π − 0 , the proportion of nulls among hypotheses with positive or negative z-scores, respectively. b b More generally, Sun and Cai”
The Bayes oracle and asymptotic optimality of multiple testing procedures under sparsity (2010, methodology): “In [43] a data-driven adaptive procedure is developed, which asymptotically attains the performance of the oracle procedure for any fixed (though unknown) proportion p of alternative hypothesis.”
Covariate‐assisted ranking and screening for large‐scale two‐sample inference (2019, methodology): “We compare the oracle and data-driven CARS procedures, respectively denoted by OR and DD, with existing methods, including the Benjamini–Hochberg procedure BH (Benjamini and Hochberg, 1995), the adaptive z-value procedure AZ (Sun and Cai, 2007) and the US procedure (Liu, 2014).”
Semi-Supervised Novelty Detection (2010, methodology): “In order to counter this type of problem, Sun and Cai (2007) suggest to estimate from the data the alterna te density and the proportion of true null hypotheses, and use these estimates directly in a plug -in likelihood ratio”
Hypothesis test for normal mixture models: The EM approach (2009, background): “They play a fundamental role in cluster analysis [Tadesse, Sha and Vannucci (2005) and Raftery and Dean (2006)], and in the study of the false discovery rate [Efron (2004), McLachlan, Bean and Ben-Tovim (2006), Sun and Cai (2007) and Cai, Jin and Low (2007)].”
Optimal detection of heterogeneous and heteroscedastic mixtures (2011, methodology): “See Sun and Cai (2007) for further details.”
Covariate powered cross‐weighted multiple testing (2017, methodology): “…where expit(u)= exp(u)∕(1+exp(u)) Falt( ⋅ |Xi = x)=Beta(𝛽(x), 1), 𝛽(x)=b0+b ⊤ x. 6Such a Bayesian, Neyman– Pearson- type procedure is motivated by the asymptotic equivalence between the frequentist FDR and the mFDR (Cai & Sun, 2009; Cai et al., 2019; Genovese & Wasserman, 2004; Sun & Cai, 2007).”
Distribution-free Multiple Testing (2016, methodology): “Most other publications that discuss oracle bounds, such as [5, 16, 24, 28, 29], operate in a setting where the statistics have the same null distribution and the same alternative distribution, and consider an oracle that provides these two distributions together with the number of false null hypoth”
Large‐scale multiple testing under dependence (2009, methodology): “The general idea in the derivation, which has been used in Genovese and Wasserman (2004), Newton et al. (2004) and Sun and Cai (2007), is first to estimate FDR for a given cut-off; then to search for the largest cut-off c such that FDR.c/ α.”
Data-driven hypothesis weighting increases detection power in genome-scale multiple testing (2016, background): “It is instructive to consider the relation between IHW and the concept of local true discovery rates. p-values are a reduction of data into one number, which typically does not contain all the important information (Table 1; [18, 19]).”
Controlling the familywise error rate with plug‐in estimator for the proportion of true null hypotheses (2009, neutral): “Sun and Cai (2007) studied suitable assumptions to obtain consistent estimators in specific mixture models.”
Global Testing and Large-Scale Multiple Testing for High-Dimensional Covariance Structures (2017, background): “Sun, W. and T. T. Cai (2007).”
Measuring reproducibility of high-throughput experiments (2011, methodology): “It is analogous to the adaptive step-up procedure of Sun and Cai (2007) for the multiple testing case.”