fwerPowerFdrPower_emp: Simulate FWER, POWER, FDR, and POWER
In mshasan/empOPW: Empirical Method of Optimal Pvalue Weighting

Description Usage Arguments Details Value Author(s) References See Also Examples

View source: R/fwerPowerFdrPower_emp.R

This function simulate Family Wise Error Rate (FWER) and corresponding Power, and False Discovery Rate (FDR) and the corresponding Power for different effect sizes

1
2
3

fwerPowerFdrPower_emp(i, simu, m, null, corr = 0, cv = 0, alpha = 0.05,
  groupSize = 100L, max.group = 5L, filterEffectVec,
  effectType = c("continuous", "binary"))

`i`	i-th filter effect
`simu`	number of replications
`m`	number of hypothesis test
`null`	proportion of the true null hypothesis
`corr`	correlation between test statistics
`cv`	determine whether the test mean effect and the filter mean effects are the same
`alpha`	significance threshold
`groupSize`	number of test statistics per group
`max.group`	maximum number of p-value groups to be used, minimum is 5
`filterEffectVec`	a vector of different effect size
`effectType`	type of effect sizes, c("continuous", "binary")

This function empirically simulate Family Wise Error Rate (FWER) and corresponding Power, and False Discovery Rate (FDR) and the corresponding Power for the different effect sizes

a matrix of 16 rows containing information about FWER, POWER, FDR, and POWER (4 rows for each)

Mohamad S. Hasan, mshasan@uga.edu

Hasan and Schliekelman (2017)

fwerPowerFdrPower

# vector of effect sizes
effectVec <- c(1, 1.5, 2)
simuVal = 2
FwerPowerFdrPower <- sapply(1:length(effectVec), fwerPowerFdrPower_emp,
             simu = simuVal, m = 10000, null = .5, corr = 0,
             cv = 0, alpha = .05, groupSize = 100, max.group = 5,
             filterEffectVec = effectVec, effectType = "continuous")