criterion {pwrFDR}

BH-FDR and Romano Criterion

Description

Compute BH-FDR step up criterion, or Romano step-down criterion

Usage

  criterion(alpha, delta, N.tests, FDP.control.method = c("BHFDR", "Romano"))

Arguments

alpha

The false discovery rate (in the BH case) or the upper bound on the probability that the FDP exceeds lambda (Romano case)

delta

If the "FDP.control.method" is set to 'Romano' then the user can set the exceedance thresh-hold for the FDP tail probability control P\{ FDP > \delta \} < \alpha. The default value is \alpha.

N.tests

The number of simultaneous hypothesis tests.

FDP.control.method

A character string specifying how the false discovery proportion (FDP) is to be controlled. You may specify the whole word or any shortened uniquely identifying truncation.
"BHFDR": the usual BH-FDR
"Romano": use Romano's method which guarantees probability less than alpha that the FDP exceeds alpha.

Value

The step down or step up criterion, which is a vector of length N.tests

Author(s)

Grant Izmirlian <izmirlian at nih dot gov>

References

Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Stat. Methodol. 1995; 57(1):289-300.

Romano J.P. and Shaikh A.M. On stepdown control of the false discovery proportion. IMS Lecture Notes–Monograph Series. 2006; 49:33-50. DOI: 10.1214/074921706000000383.

Examples

  library(pwrFDR)

  crit.b <- criterion(N.tests=1000, alpha=0.15, FDP.control.method="BHFDR")
  crit.r <- criterion(N.tests=1000, alpha=0.15, FDP.control.method="Romano")
  crit.r.17 <- criterion(N.tests=1000, alpha=0.15, delta=0.17, FDP.control.method="Romano")
  matplot(1:1000, cbind(crit.b, crit.r, crit.r.17), type="l", lty=1, col=2:4)