findMaxD2 {edgeR}R Documentation

Maximizes the negative binomial likelihood

Description

Maximizes the negative binomial likelihood (a weighted version using the common likelihood given weight alpha) for each tag

Usage

 
findMaxD2(object, alpha = 0.5, grid = TRUE, tol = 1e-05, n.iter = 10, grid.length = 200)

Arguments

object list containing the raw data with elements data (table of counts), group (vector indicating group) and lib.size (vector of library sizes)
alpha weight given to common likelihood, set to 0 for individual estimates or large (e.g. 100) for common likelihood
grid logical, whether to use a grid search (default = TRUE); if FALSE use Newton-Rhapson steps
tol if grid=FALSE, tolerance for Newton-Rhapson iterations
n.iter if grid=FALSE, number of Newton-Rhapson iterations
grid.length length of the grid over which to maximize; default 200

Value

vector of the values of delta that maximize the negative binomial likelihood for each tag (where delta = phi / (phi+1) and phi is the overdispersion parameter)

Author(s)

Mark Robinson, Davis McCarthy

Examples

y<-matrix(rnbinom(1000,mu=10,size=2),ncol=4)
d<-DGEList(data=y,group=c(1,1,2,2),lib.size=c(1000:1003))
cml1<-findMaxD2(d,alpha=10)
cml2<-findMaxD2(d,alpha=0)

[Package edgeR version 1.2.4 Index]