nbinomUnitDeviance: Negative Binomial Unit Deviance

In edgeR: Empirical Analysis of Digital Gene Expression Data in R

Description

Compute unit deviances for the negative binomial distribution.

Usage

nbinomUnitDeviance(y, mean, dispersion = 0)

Arguments

y	numeric vector or matrix containing negative binomial counts.
mean	numeric vector or matrix of expected values. If a matrix, then of same dimensions as y.
dispersion	numeric vector or matrix of negative binomial dispersions. Can be a scalar, a vector of length nrow(y) or a matrix of same dimensions as y.

Details

The unit deviance of the negative binomial distribution is a measure of the distance between y and mean. If mean and dispersion are the true mean and dispersion of the negative binomial distribution, then the unit deviance follows an approximate chisquare distribution on 1 degree of freedom.

This function computes the unit deviance for each y observation. Care is taken to ensure accurate computation in limiting cases when the dispersion is near zero or mean*dispersion is very large.

Value

Numeric vector or matrix of the same size as y containing unit deviances.

Author(s)

Gordon Smyth, Yunshun Chen, Aaron Lun. C++ code by Aaron Lun.

References

Jorgensen, B. (2013). Generalized linear models. Encyclopedia of Environmetrics 3, Wiley. http://onlinelibrary.wiley.com/doi/10.1002/9780470057339.vag010.pub2/abstract.

McCarthy, DJ, Chen, Y, Smyth, GK (2012). Differential expression analysis of multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids Research 40, 4288-4297. https://doi.org/10.1093/nar/gks042

Examples

y <- 1:4
names(y) <- letters[1:4]
nbinomUnitDeviance(y,mean=2.5,dispersion=0.2)

Example output

Loading required package: limma
        a         b         c         d 
0.8451411 0.0733260 0.0613130 0.4782410 

edgeR documentation built on Jan. 16, 2021, 2:03 a.m.