fit-correlated-method: Fit method for correlated phase-type frailty models

In jorgeyslas/phfrailty: Phase-type frailty models

Fit method for correlated phase-type frailty models

Description

Fit method for correlated phase-type frailty models

Usage

## S4 method for signature 'correlated'
fit(
  x,
  y,
  X = numeric(0),
  stepsEM = 100,
  stepsPH = 50,
  initialpoint1 = 1e-04,
  truncationpoint1 = 8,
  delta1 = 0.1,
  initialpoint2 = 1e-04,
  truncationpoint2 = 8,
  delta2 = 0.1,
  maxit = 100,
  reltol = 1e-08,
  every = 100
)

Arguments

x	An object of class correlated.
y	Matrix or data.
X	A matrix of covariates.
stepsEM	Number of EM steps to be performed.
stepsPH	Number of EM steps for the phase-type component at each iteration of the global EM.
initialpoint1	Initial value for discretization of first marginal density.
truncationpoint1	Ultimate value for discretization of first marginal density.
delta1	Size of interval for discretization of first marginal.
initialpoint2	Initial value for discretization of second marginal density.
truncationpoint2	Ultimate value for discretization of second marginal density.
delta2	Size of interval for discretization of second marginal.
maxit	Maximum number of iterations when optimizing g function.
reltol	Relative tolerance when optimizing g function.
every	Number of iterations between likelihood display updates.

Value

An object of class correlated.

Examples

bivph_obj <- bivphasetype(dimensions = c(3, 3))
cobj <- correlated(bivph_obj, bhaz1 = "weibull", bhaz_pars1 = 3, bhaz2 = "weibull", bhaz_pars2 = 2)
data <- sim(cobj, n = 100)
fit(cobj, data, stepsEM = 5, stepsPH = 5, every = 1)

jorgeyslas/phfrailty documentation built on April 17, 2025, 4:11 p.m.