glmgam.fit {statmod}R Documentation

Fit Generalized Linear Model by Fisher Scoring with Levenberg Damping

Description

Fit a generalized linear model with secure convergence. Provided for gamma glm with identity links or negative binomial glm with log-links.

Usage

glmgam.fit(X, y, start=NULL, tol=1e-6, maxit=50, trace=FALSE)
glmnb.fit(X, y, dispersion, offset=0, start=NULL, tol=1e-6, maxit=50, trace=FALSE)

Arguments

X design matrix, assumed to be of full column rank. Missing values not allowed.
y numeric vector of responses. Negative or missing values not allowed.
dispersion over-dispersion parameter for negative binomial.
offset offset vector for linear model
start numeric vector of starting values for the regression coefficients
tol small positive numeric value giving convergence tolerance
maxit maximum number of iterations allowed
trace logical value. If TRUE then output diagnostic information at each iteration.

Details

These functions implement a modified Fisher scoring algorithm for generalized linear models, similar to the Levenberg-Marquardt algorithm for nonlinear least squares. The Levenberg-Marquardt modification checks for a reduction in the deviance at each step, and avoids the possibility of divergence. The result is a very secure algorithm that converges for almost all datasets.

glmgam.fit is in principle similar to glm.fit(X,y,family=Gamma(link="identity")) but with much more secure convergence. This function is used by mixedModel2Fit.

glmnb.fit is in principle similar to glm.fit(X,y,family=negative.binomial(link="log",theta=1/dispersion)) but with more secure convergence.

Value

List with the following components:
coefficients numeric vector of regression coefficients
fitted numeric vector of fitted values
deviance residual deviance
iter number of iterations used to convergence. If convergence was not achieved then iter is set to maxit+1.

Author(s)

Gordon Smyth and Yunshun Chen

Examples

y <- rgamma(10,shape=5)
X <- cbind(1,1:10)
fit <- glmgam.fit(X,y,trace=TRUE)

[Package statmod version 1.4.10 Index]