match_pvals {DiscreteFWER} | R Documentation |
Matching Raw P-Values with Supports
Description
Constructs the observed p-values from the raw observed p-values, by rounding
them to their nearest neighbour, matching with the supports of their
respective CDFs (as in function p.discrete.adjust()
of package
discreteMTP
, which is no longer available on CRAN).
Note: This is an internal function and has to be called directly via
:::
, i.e. DiscreteFWER:::match_pvals()
.
Usage
match_pvals(test_results, pCDFlist, pCDFlist_indices = NULL)
Arguments
test_results |
either a numeric vector with |
pCDFlist |
list of the supports of the CDFs of the |
pCDFlist_indices |
list of numeric vectors containing the test indices that indicate to which raw |
Details
Well computed raw p-values should already belong to their respective CDF
support. So this function is called at the beginning of
discrete_FWER.default()
and its wrappers, just in case raw p-values may
be biased.
For each raw p-value that needs to be rounded, a warning is issued.
Value
A vector where each raw p-value has been replaced by its nearest neighbour, if necessary.