emfrail_dist: Distribution parameters for emfrail

Description

Distribution parameters for emfrail

Usage

emfrail_dist(dist = "gamma", theta = 2, pvfm = -1/2,
  left_truncation = FALSE, basehaz = "breslow")

Arguments

dist

One of 'gamma', 'stable' or 'pvf'.

theta

A starting value for the 'outer' maximization with respect to the frailty parameter \(\theta\). Must be >0.

pvfm

Only relevant if dist = 'pvf' is used. It determines which PVF distribution should be used. Must be larger than -1 and not equal to 0.

left_truncation

Logical. Whether the data set represents left truncated survival times.

basehaz

A character string which determines how the baseline hazard is calculated. The default is "breslow", but other possible options are "weibull", "exponential" "gaussian", "logistic", "lognormal" or "loglogistic".

Value

An object of the type emfrail_dist, which is mostly used to denote the supported frailty distributions in a consistent way.

Details

The theta argument must be positive. In the case of gamma or PVF, this is the inverse of the frailty variance, i.e. the larger the theta is, the closer the model is to a Cox model. When dist = "pvf" and pvfm = -0.5, the inverse Gaussian distribution is obtained. For the positive stable distribution, the \(\gamma\) parameter of the Laplace transform is \(\theta / (1 + \theta)\), with the \(alpha\) parameter fixed to 1.

Examples

Run this code

# NOT RUN {
emfrail_dist()
# Compound Poisson distribution:
emfrail_dist(dist = 'pvf', theta = 1.5, pvfm = 0.5)
# Inverse Gaussian distribution:
emfrail_dist(dist = 'pvf')
# }