EBfit {SharedHT2}R Documentation

Extractor function

Description

Extract the fit component from an object of class fit.n.data

Usage

  EBfit(object)

Arguments

object An object of class fit.n.data returned by EB.Anova

Value

An object of class EBfit containing the following components:

coefficients a numerical vector of length d*(d+1)/2 + 1 containing the model parameters. The first is the logged shape parameter of the inverse Wishart, and the remaining d*(d+1)/2 are related to the its rate matrix parameter, which is symmetric and d by d. The first d of these are the logged diagonal elements and the remaining d*(d-1)/2 are the off-diagonal elements.
variance the estimated d*(d+1)/2 + 1 by d*(d+1)/2 + 1 covariance matrix of the model coefficients
gradient the value of the gradient vector at the MLE–hopefully very close to zero.
nu the shape parameter of the inverse Wishart = exp(coefficients[1])
Lambda the rate parameter of the inverse Wishart, a positive definite, symmeteric d by d matrix.
likelihood the log likelihood...actually positive (like I said, big spike)
fail, fncount, grcount, mask details from the optimization routine
usegr whether the gradient method was used in the optimization
call the original call to EB.Anova

Author(s)

Grant Izmirlian izmirlian@nih.gov

See Also

EB.Anova, EBfit, SimAffyDat, TopGenes, SimNorm.IG, SimMVN.IW, SimMVN.mxIW, SimOneNorm.IG, SimOneMVN.IW, SimOneMVN.mxIW

Examples

#  See examples under ?EB.Anova  

[Package SharedHT2 version 2.0 Index]