R: Repeated Events Models with Frailty or Serial Dependence

kalsurv {event}

R Documentation

Repeated Events Models with Frailty or Serial Dependence

Description

kalsurv is designed to handle event history models with time-varying covariates. The distributions have two extra parameters as compared to the functions specified by intensity and are generally longer tailed than those distributions. Dependence of inter-event times can be through gamma frailties (a type of random effect), with or without autoregression, or several kinds of serial dependence by updating, as in Kalman filtering.

By default, a gamma mixture of the distribution specified in intensity is used, as the conditional distribution in the serial dependence models, and as a symmetric multivariate (random effect) model for frailty dependence. For example, with a Weibull intensity and frailty dependence, this yields a multivariate Burr distribution and with Markov or serial dependence, univariate Burr conditional distributions.

If a value for pfamily is used, the gamma mixture is replaced by a power variance family mixture.

Nonlinear regression models can be supplied as formulae where parameters are unknowns in which case factor variables cannot be used and parameters must be scalars. (See finterp.)

Marginal and individual profiles can be plotted using mprofile and iprofile and residuals with plot.residuals.

Usage

kalsurv(response, intensity="exponential", distribution="Pareto",
	depend="independence", update="Markov", mu=NULL, shape=NULL,
	renewal=TRUE, density=FALSE, censor=NULL, delta=NULL, ccov=NULL,
	tvcov=NULL, preg=NULL, ptvc=NULL, pbirth=NULL,
	pintercept=NULL, pshape=NULL, pinitial=1, pdepend=NULL,
	pfamily=NULL, envir=parent.frame(), print.level=0,
	ndigit=10, gradtol=0.00001, steptol=0.00001, iterlim=100,
	fscale=1, typsize=abs(p), stepmax=10*sqrt(p%*%p))

Arguments

`response`	A list of vectors with times between events for each individual, one matrix or dataframe of such times if all individuals have the same number of events, or an object of class, `response` (created by `restovec`) or `repeated` (created by `rmna` or `lvna`). If the `repeated` data object contains more than one response variable, give that object in `envir` and give the name of the response variable to be used here.
`intensity`	The form of intensity function to be put in the distribution given by dist. Choices are exponential, Weibull, gamma, log normal, log logistic, log Cauchy, log Student, and gen(eralized) logistic.
`distribution`	The outer distribution. Choices are Pareto, gamma, and Weibull.
`depend`	Type of dependence. Choices are `independence`, `frailty`, and `serial`.
`update`	Type of update for serial dependence. Choices are `Markov`, `elapsed Markov`, `serial`, `event`, `cumulated`, `count`, and `kalman`. With `frailty` dependence, weighting by length of observation time may be specified by setting update to `time`.
`mu`	A regression function for the location parameter or a formula beginning with ~, specifying either a linear regression function in the Wilkinson and Rogers notation or a general function with named unknown parameters. Give the initial estimates in `preg` if there are no time-varying covariates and in `ptvc` if there are.
`shape`	A regression function for the shape parameter or a formula beginning with ~, specifying either a linear regression function in the Wilkinson and Rogers notation or a general function with named unknown parameters. It must yield one value per observation.
`renewal`	IF TRUE, a renewal process is modelled, with time reinitialized after each event. Otherwise, time is cumulated from the origin of observations.
`density`	If TRUE, the density of the function specified in `intensity` is used instead of the intensity.
`censor`	A vector of the same length as the number of individuals containing a binary indicator, with a one indicating that the last time period in the series terminated with an event and zero that it was censored. For independence and frailty models, where response is matrix, censor may also be a matrix of the same size. Ignored if response has class, `response` or `repeated`.
`delta`	Scalar or vector giving the unit of measurement for each response value, set to unity by default. For example, if a response is measured to two decimals, delta=0.01. If the response has been pretransformed, this must be multiplied by the Jacobian. This transformation cannot contain unknown parameters. For example, with a log transformation, `delta=1/y`. (The delta values for the censored response are ignored.) Ignored if response has class, `response` or `repeated`.
`ccov`	A vector or matrix containing time-constant baseline covariates with one entry per individual, a model formula using vectors of the same size, or an object of class, `tccov` (created by `tcctomat`). If response has class, `repeated`, the covariates must be supplied as a Wilkinson and Rogers formula unless none are to be used or `mu` is given.
`tvcov`	A list of matrices with time-varying covariate values, observed at the event times in `response`, for each individual (one column per variable), one matrix or dataframe of such covariate values, or an object of class, `tvcov` (created by `tvctomat`). If response has class, `repeated`, the covariates must be supplied as a Wilkinson and Rogers formula unless none are to be used or `mu` is given.
`preg`	Initial parameter estimates for the regression model: intercept plus one for each covariate in `ccov`. If `mu` is a formula or function, the parameter estimates must be given here only if there are no time-varying covariates. If `mu` is a formula with unknown parameters, their estimates must be supplied either in their order of appearance in the expression or in a named list.
`ptvc`	Initial parameter estimates for the coefficients of the time-varying covariates, as many as in `tvcov`. If `mu` is a formula or function, the parameter estimates must be given here if there are time-varying covariates present.
`pbirth`	If supplied, this is the initial estimate for the coefficient of the birth model.
`pintercept`	The initial estimate of the intercept for the generalized logistic intensity.
`pshape`	An initial estimate for the shape parameter of the intensity (except exponential intensity). If `shape` is a function or formula, the corresponding initial estimates. If `shape` is a formula with unknown parameters, their estimates must be supplied either in their order of appearance in the expression or in a named list.
`pinitial`	An initial estimate for the initial parameter. In `frailty` dependence, this is the frailty parameter.
`pdepend`	An initial estimate for the serial dependence parameter. For `frailty` dependence, if a value is given here, an autoregression is fitted as well as the frailty.
`pfamily`	An optional initial estimate for the second parameter of a two-parameter power variance family mixture instead of the default gamma mixture. This yields a gamma mixture as `family -> 0`, an inverse Gauss mixture for `family = 0.5`, and a compound distribution of a Poisson-distributed number of gamma distributions for `-1 < family < 0`.
`envir`	Environment in which model formulae are to be interpreted or a data object of class, `repeated`, `tccov`, or `tvcov`; the name of the response variable should be given in `response`. If `response` has class `repeated`, it is used as the environment.
`print.level`	`nlm` control options.
`ndigit`	`nlm` control options.
`gradtol`	`nlm` control options.
`steptol`	`nlm` control options.
`iterlim`	`nlm` control options.
`fscale`	`nlm` control options.
`typsize`	`nlm` control options.
`stepmax`	`nlm` control options.

Value

A list of classes kalsurv and recursive is returned.

Author(s)

J.K. Lindsey

Examples

treat <- c(0,0,1,1)
tr <- tcctomat(treat)
cens <- matrix(rbinom(20,1,0.9),ncol=5)
times <- # matrix(rweibull(20,2,1+3*rep(treat,5)),ncol=5)
	matrix(c(1.36,0.18,0.84,0.65,1.44,1.79,1.04,0.43,1.35,1.63,2.15,1.15,
		1.21,5.46,1.58,3.44,4.40,2.75,4.78,2.44),ncol=5,byrow=TRUE)
times <- restovec(times, censor=cens)
reps <- rmna(times, ccov=tr)
# exponential intensity model with independence
kalsurv(times, pinitial=0.5, preg=1, dep="independence",
	intensity="exponential")
# Weibull intensity model with independence
kalsurv(times, pinitial=0.5, preg=1, pshape=1, dep="independence",
	intensity="Weibull")
# same model with serial update
kalsurv(times, pinitial=0.5, pdep=0.1, preg=1, pshape=1, dep="serial",
	intensity="Weibull")
# try power variance family instead of gamma distribution for mixture
kalsurv(times, pinitial=0.5, pdep=0.1, preg=1, pshape=1, dep="serial",
	intensity="Weibull", pfamily=0.1)
# treatment effect with log link
kalsurv(times, pinitial=0.5, preg=c(1,0), pshape=1, intensity="Weibull",
	ccov=treat)
# or equivalently
kalsurv(times, mu=~exp(a+b*treat), pinitial=0.1, preg=c(1,0), pshape=1,
	intensity="Weibull", envir=reps)
# with identity link instead
kalsurv(times, mu=~treat, pinitial=0.5, preg=c(1,0), pshape=1,
	intensity="Weibull")
# or equivalently
kalsurv(times, mu=~a+b*treat, pinitial=0.5, preg=c(1,0), pshape=1,
	intensity="Weibull", envir=reps)
# add the birth model
kalsurv(times, pinitial=0.5, preg=c(1,0), pshape=1,
	intensity="Weibull", ccov=treat, pbirth=0)
# try frailty dependence
kalsurv(times, pinitial=0.5, preg=c(1,0), pshape=1, dep="frailty",
	intensity="Weibull", ccov=treat)
# add autoregression
kalsurv(times, pinitial=0.5, preg=c(1,0), pshape=1, dep="frailty",
	pdep=0.1, intensity="Weibull", ccov=treat)
# switch to gamma intensity model
kalsurv(times, pinitial=0.5, preg=c(1,0), pshape=1, intensity="gamma",
	ccov=treat)

[Package event version 1.1.1 Index]