simulate_fPi0_t_tests: simulation of t-tests with functional proportion
In StoreyLab/fFDR: Functional FDR for multiple hypothesis testing

Unlike simulateTTests, this simulation fixes the sample size of each t-test but models the likelihood that a null hypothesis is true as a function of a random variable z0.

1	simulate_fPi0_t_tests(shape = "Monotonic", m = 2500, n = 30, ...)

`shape`	True shape of pi0 relative to z0: either "Monotonic", "Asymptotic", "Symmetric", "2 Step" or "3 Step"
`m`	Number of hypotheses
`n`	Sample size of each test
`...`	Additional parameters passed on to simulateTTests

First, the random variable z0 is generated from a standard Normal distribution. Then the likelihood of a true null hypothesis pi0 is modelled as a function of z0.

z is calculated as rank(z0) / length(z0).

StoreyLab/fFDR documentation built on March 8, 2021, 10:14 p.m.