eha (version 1.0)

wfunk: Loglihood function of a Weibull regression

Description

Calculates minus the log likelihood function and its first and second order derivatives for data from a Weibull regression model. Is called by weibreg.

Usage

wfunk(beta = NULL, lambda, p, X = NULL, Y, offset = rep(0, length(Y)),
ord = 2, pfixed = FALSE, trace = TRUE)

Arguments

beta

Regression parameters

lambda

The scale paramater

p

The shape parameter

X

The design (covariate) matrix.

Y

The response, a survival object.

offset

Offset.

ord

ord = 0 means only loglihood, 1 means score vector as well, 2 loglihood, score and hessian.

pfixed

Logical, if TRUE the shape parameter is regarded as a known constant in the calculations.

trace

If TRUE, more verbose output.

Value

• A list with components
• f
  The log likelihood. Present if ord >= 0
• fp
  The score vector. Present if ord >= 1
• fpp
  The negative of the hessian. Present if ord >= 2

Details

Note that the function returns log likelihood, score vector and minus hessian, i.e. the observed information. The model is $$h(t;p,\lambda ,\beta ,z)=p/\lambda (t/\lambda {)}^{(p-1)}\exp (-(t/\lambda {)}^p)\exp (z\beta )$$ This is in correspondence with dweibull.

See Also

weibreg