R: Loglihood function (partial likelihood) of a Cox regression

coxfunk {eha}

R Documentation

Loglihood function (partial likelihood) of a Cox regression

Description

Calculates minus the log likelihood function and its first and second order derivatives for data from a Cox regression model. It is used by coxreg if the argument coxph = FALSE

Usage

coxfunk(beta, X, offset, rs, what = 2)

Arguments

`beta`	Regression parameters
`X`	The design (covariate) matrix.
`offset`	Offset.
`rs`	Risk set created by `risksets(..., collate_sets = TRUE)`
`what`	what = 0 means only loglihood, 1 means score vector as well, 2 loglihood, score and hessian.

Details

Note that the function returns log likelihood, score vector and minus hessian, i.e. the observed information. The model is

Value

A list with components

`loglik`	The log likelihood.
`dloglik`	The score vector. Nonzero if `what >= 1`
`d2loglik`	The hessian. Nonzero if `ord >= 2`

Author(s)

Göran Broström