kalord {ordinal}R Documentation

Repeated Measurements Models for Ordinal Variables with Frailty or Serial Dependence

Description

kalord is designed to handle repeated measurements models with time-varying covariates. The distributions have two extra parameters as compared to the parameterization of the logistic distribution specified by distribution. Dependence among observations on a unit can be through gamma frailties (a type of random effect) or serial dependence over time.

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, individual and predicted profiles can be plotted using moprofile, ioprofile and poprofile.

If the responses on a unit are clustered, not longitudinal, use the frailty dependence.

Usage

kalord(response,times=NULL,distribution="multinomial",
       depend="independence",mu=NULL,ccov=NULL,tvcov=NULL,torder=0,
       interaction=NULL,preg=NULL,ptvc=NULL,pinitial=1,pdepend=NULL,
       envir=sys.frame(sys.parent()),optimize=T,print.level=0,
       ndigit=10,gradtol=0.00001,steptol=0.00001,fscale=1,
       iterlim=100,typsiz=abs(p),stepmax=10*sqrt(p

Arguments

response A list of two column matrices with responses and corresponding times for each individual, one matrix or dataframe of response values, 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.
times When response is a matrix, a vector of possibly unequally spaced times when they are the same for all individuals or a matrix of times. Not necessary if equally spaced. Ignored if response has class, response or repeated.
distribution Specifies the parameterization of the logistic distribution to put in the Pareto distribution. Choices are binary, multinomial, continuation-ratio, and proportional-odds.
depend Type of dependence. Choices are independence, Markov, serial, and frailty.
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. The regression function must not contain intercepts. Give the initial estimates in preg or in ptvc.
ccov A vector or matrix containing time-constant baseline covariates with one row 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 a time-varying covariate is observed at arbitrary time, gettvc can be used to find the most recent values for each response and create a suitable list. 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.
torder The order of the polynomial in time to be fitted.
interaction Vector of length equal to the number of time-constant covariates, giving the levels of interactions between them and the polynomial in time in the linear model.
preg Initial parameter estimates for the regression model: intercept, one for each covariate in ccov, and torder plus sum(interaction). 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.
pinitial An initial estimate for the initial parameter, if set to NULL this parameter will be fixed at zero. (With frailty dependence, this is the frailty parameter.)
pdepend An initial estimate for the serial dependence parameter.
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.
optimize If set to TRUE then nlm is used to perform the numerical optimization of the likelihood function, otherwise if set to FALSE no optimization is performed.
others Arguments controlling nlm.

Value

A list of classes kalordinal and recursive is returned.

Author(s)

P.J. Lindsey

See Also

finterp, gettvc, ioprofile, lvna, moprofile, plot.ordinal, poprofile, restovec, rmna, tcctomat, tvctomat.

Examples

library(ordinal)

#
# Binary data
#
data(cardiac.indiv)

y <- restovec(cardiac.indiv[,1:4],type="ordinal")

cov <- tcctomat(as.matrix(cardiac.indiv[,5:10]))

w <- rmna(y,ccov=cov)

rm(cardiac.indiv,y,cov)

# Time-constant covariate.
kalord(w,distribution="binary",ccov=~age,preg=c(3.9507,-0.0308),pinit=NULL)

# Time-varying covariate.
kalord(w,distribution="binary",tvcov=~times,preg=1.832,ptvc=0.0573,pinit=NULL)

# Time-constant and time-varying covariate.
kalord(w,distribution="binary",mu=~age+ren+cop+dia+sex+pmi+times,
       ptvc=c(3.888,-0.0289,-0.642,-0.366,-0.314,-0.154,-0.114,0.057),pinit=NULL)

# Time-constant and time-varying covariate with a frailty dependence.
kalord(w,distribution="binary",mu=~age+ren+cop+dia+sex+pmi+times,
       ptvc=c(4.43391,-0.03128,-0.62439,-0.37596,-0.33064,-0.17095,-0.12216,-0.09096),
       pinit=0.1196,dep="frailty")

rm(w)

#
# Ordinal data
#
data(tmi2)

y <- restovec(tmi2[,1:4],type="ordinal")

cov <- tcctomat(tmi2[,5],name="distance")

w <- rmna(y,ccov=cov)

rm(tmi2,y,cov)

# Continuation-ratio model with time-constant covariate with a serial dependence.
kalord(w,distribution="continuation-ratio",ccov=~distance,preg=c(-1.907,7.7,-0.162),
       pinit=2.55,pdep=0.328,dep="serial")

# Proportional-odds model with time-constant covariate with a Markov dependence.
kalord(w,distribution="proportional-odds",ccov=~distance,preg=c(-1.89,11.652,-0.199),
        pinit=3.111,pdep=0.217,dep="Markov")

rm(w)

[Package ordinal version 0.3 Index]