cFDR: cFDR: a package for estimating conditional false discovery...
In chr1swallace/cFDR:

Description Usage Arguments Value Author(s)

The main functions are cFDR and FDR

Calculate cFDR given vectors of *independent* test and conditional Z scores

1
2
3

cFDR(Z, Zc, eps = 0.01, alpha = seq(0.01, 0.1, by = 0.005),
  do.optimise = FALSE, method = c("empirical", "unif", "means", "vars",
  "both"), ...)

`Z`	vector of Z scores
`Zc`	vector of Z scores for the conditional tests
`eps`	tolerance for FDR control to alpha
`alpha`	vector of thresholds. overall cFDR will be assessed if all tests with cFDR<alpha were rejected
`do.optimise`	logical, default FALSE. if true, and a single value of alpha is supplied, the rejection threshold will be optimised to control the overall FDR at alpha.
`...`	other arguments passed to cFDR.tune

a named list with three components. alpha is the threshold for rejection, est.cFDR is the estimated FDR at this threshold, cFDR is a vector of conditional FDRs in rectangles defined the corresponding absolute Z and Zc values.

Chris Wallace

chr1swallace/cFDR documentation built on May 13, 2019, 6:18 p.m.