
False discovery rate control for multiple testing based on pvalues with càdlàg distribution functions
For multiple testing based on pvalues with càdlàg distribution function...
read it

Gaining power in multiple testing of interval hypotheses via conditionalization
In this paper we introduce a novel procedure for improving multiple test...
read it

Decentralized Nonparametric Multiple Testing
Consider a big data multiple testing task, where, due to storage and com...
read it

Estimators of the proportion of false null hypotheses: I "universal construction via LebesgueStieltjes integral equations and uniform consistency under independence"
The proportion of false null hypotheses is a very important quantity in ...
read it

Testing for Outliers with Conformal pvalues
This paper studies the construction of pvalues for nonparametric outlie...
read it

ADDIS: adaptive algorithms for online FDR control with conservative nulls
Major internet companies routinely perform tens of thousands of A/B test...
read it

Accurate and Efficient Estimation of Small Pvalues with the CrossEntropy Method: Applications in Genomic Data Analysis
Small pvalues are often required to be accurately estimated in large sc...
read it
Improved qvalues for discrete uniform and homogeneous tests: a comparative study
Large scale discrete uniform and homogeneous Pvalues often arise in applications with multiple testing. For example, this occurs in genome wide association studies whenever a nonparametric onesample (or twosample) test is applied throughout the gene loci. In this paper we consider qvalues for such scenarios based on several existing estimators for the proportion of true null hypothesis, π_0, which take the discreteness of the Pvalues into account. The theoretical guarantees of the several approaches with respect to the estimation of π_0 and the false discovery rate control are reviewed. The performance of the discrete qvalues is investigated through intensive Monte Carlo simulations, including location, scale and omnibus nonparametric tests, and possibly dependent Pvalues. The methods are applied to genetic and financial data for illustration purposes too. Since the particular estimator of π_0 used to compute the qvalues may influence the power, relative advantages and disadvantages of the reviewed procedures are discussed. Practical recommendations are given.
READ FULL TEXT
Comments
There are no comments yet.