multinbmod (version 1.0)

multinb.fit: Multivariate negative binomial model with robust estimation of regression coefficients

Description

This function is called by "multinbmod", but it can also be called directly

Usage

multinb.fit(y, x, offset=1, id, start.par, control=list())

Arguments

y

Response vector.

x

Design matrix of covariates.

offset

Optional vector of offset values.

id

Variable indicating which subjects are correlated.

start.par

Vector of starting values for the parameters in the linear predictor (defaults to zero) and the overdispersion parameter (default to 0.5).

control

A list of parameters that control the convergence criteria. See "nlminb" for details.

Value

estimated regression coefficients

se from model

Estimated standard errors of regression coefficients.

robust se

Robust estimate of standard errors of regression coefficients.

t-values

Robust t-values.

covariance of beta estimates from model

Estimated covariance of estimated regression parameters.

robust covariance of beta estimates

Robust estimate of covariance of estimated regression coefficients

estimated phi

ML estimate of overdisperision parameter.

se(phi)

Its standard error.

-2 x log-likelihood

converged?

Logical.

iterations

Number of iterations required for convergence.

References

Solis-Trapala, I.L. and Farewell, V.T. (2005) Regression analysis of overdispersed correlated count data with subject specific covariates. Statistics in Medicine, 24: 2557-2575.

See Also

multinbmod

Examples

id <- factor(rep(1:20, rep(5, 20)))
y <- rnbinom(100, mu = rexp(100,1)+rep(rexp(20,.3),rep(5,20)),size=2.5)
x<-rbinom(100,1,.5)
dat <- data.frame(y = y, x = x, id = id)
multinb.fit(y,cbind(1,x),id=id)