ctsem (version 2.6.4)

ctFitR: Fit ctsem models with Kalman filter R code -- slower!

Description

Fit ctsem models with Kalman filter R code -- slower!

Usage

ctFitR(datalong, ctmodel, ...)

Arguments

datalong

long data

ctmodel

ctsem model of type 'omx'

...

arguments to pass to ctsem::Kalman

Value

matrix of estimates and standard errors

Examples

# NOT RUN {
Tpoints<-250
n.manifest=5
gm<-ctModel(type='omx',n.latent=1,n.manifest=n.manifest,Tpoints=Tpoints,
LAMBDA=matrix(rep(1,n.manifest),ncol=1),
  DRIFT=diag(-.3,1),
  T0VAR=diag(1),
  MANIFESTMEANS=matrix(0,nrow=n.manifest),
  MANIFESTVAR=t(chol(diag(.001,n.manifest))),
  DIFFUSION=t(chol(diag(1,1))))

cd<-ctGenerate(gm,n.subjects=1,burnin=300,wide=FALSE)

gm2=gm
gm2$DRIFT[1]='drift11'

cfit=ctFit(dat=cd,ctmodelobj=gm2,dataform='long')
cfit$mxobj$DRIFT$values

rfit=ctFitR(datalong=cd,ctmodel=gm2)
rfit
# }