null.sn.fit: null.sn.fit

In npetraco/fdrID: Computes local false discovery rate probabilities using JAGS, Stan and code from Efron, Turnbull and Narasimhan locfdr

Description

Skewed normal fit for null or transformed null similarity scores.

Usage

null.sn.fit(score.null.vec, standardizeQ = FALSE, plotQ = FALSE)

Arguments

score.null.vec	Vector of null (non-match) similarity scores or transformed similarity scores
standardizeQ	Whether or not to standardize the null scores
plotQ	Diagnostic plots?

Details

The skewed normal distribution is a flexible three parameter distribution which often gives a good fit to the null log similarity scores when these scores are on a scale of 0 to 1. Other than standardization, this function assumes that if the user wants the scores to be transformed, they've transformed them. This routine calls the selm function from the sn package.

Value

list with the fitted parameters, fit info and chi-square goodness of fit test results

References

XXXX

Examples

XXXX

npetraco/fdrID documentation built on May 23, 2019, 9:33 p.m.