intern.estim: Internal function

In rauschenberger/globalSeq: Global Test for Counts

Description

This functions estimates the parameters of the negative binomial distribution by maximum likelihood. It is called by the functions omnibus and proprius.

Usage

intern.estim(y, offset = NULL)

Arguments

y	random variable: numeric vector of length n
offset	numeric vector of length n

Details

We assume the negative binomial distribution y_i ~ NB(mu,phi), where the samples are indexed by i (i=1,...,n). Our parametrisation leads to E[y]= mu and Var[y]= mu + phi*mu^2. With the an offset the model becomes y_i ~ NB(a_i*mu,phi), where the a_i are known.

Value

The function returns a list of numeric vectors.

References

A Rauschenberger, MA Jonker, MA van de Wiel, and RX Menezes (2016). "Testing for association between RNA-Seq and high-dimensional data", BMC Bioinformatics. 17:118. html pdf (open access)

See Also

This is an internal function. The user functions are cursus, omnibus, and proprius.

Examples

set.seed(1)
y <- rnbinom(n=1000,mu=10,size=1/0.2)
intern.estim(y)

rauschenberger/globalSeq documentation built on May 19, 2020, 4:09 a.m.