ctsem (version 1.1.6)

ctGenerate: Simulate continuous time data

Description

This function generates data according to the specified ctsem model object. Not all T0 matrices are included at present, safest to use a high burnin (where 'high' is sufficient for the process to forget starting values)

Usage

ctGenerate(ctmodelobj, n.subjects = 1000, burnin = 0, dT = 1, asymptotes = FALSE)

Arguments

ctmodelobj

ctsem model object from ctModel.

n.subjects

Number of subjects to output.

burnin

Number of initial time points to discard (to simulate stationary data)

dT

Time interval (delta T) to use, defaults to 1.

asymptotes

Are the parameters provided asymptotic paramters, or the regular continuous time parameters?

Details

TRAITTDPREDCOV and TIPREDCOV matrices are not accurately accounted for, at present.

Examples

#generate data for 2 process model, each process measured by noisy indicator, 
#stable individual differences in process levels.

generatingModel<-ctModel(Tpoints=8,n.latent=2,n.TDpred=0,n.TIpred=0,n.manifest=2,
 MANIFESTVAR=diag(.1,2),
 LAMBDA=diag(1,2),
 DRIFT=matrix(c(-.2,-.05,-.1,-.1),nrow=2),
 TRAITVAR=matrix(c(.5,.2,0,.8),nrow=2),
 DIFFUSION=matrix(c(1,.2,0,4),2),
 CINT=matrix(c(1,0),nrow=2),
 T0MEANS=matrix(0,ncol=1,nrow=2),
 T0VAR=diag(1,2))

data<-ctGenerate(generatingModel,n.subjects=150,burnin=500)

## Further examples set to 'dontrun' because they take longer than 5s. 

## Not run: 
# ctIndplot(data,n.manifest=2,Tpoints=4,n.subjects=10)
# 
# model<-ctModel(Tpoints=8, TRAITVAR='auto', n.latent=2,
#  n.manifest=2, LAMBDA=diag(2))
# 
# checkf<-ctFit(data,model,stationary=c('T0VAR','T0MEANS'))
# summary(checkf,verbose=TRUE)
#  
#  
# #### with 2 process from 4 indicators, latent trait, TDpred and TIpred 
# Tpoints=8
# n.latent=2
# n.manifest=4
# n.TDpred=1
# n.TIpred=1
# 
# generatingModel<-ctModel(Tpoints=Tpoints,n.latent=n.latent,
#  n.TDpred=n.TDpred,n.TIpred=n.TIpred,n.manifest=n.manifest,
#  LAMBDA=matrix(c(1,.4,.8,0,0,0,0,1),nrow=n.manifest,ncol=n.latent),
#  MANIFESTVAR=diag(c(.2),n.manifest),
#  MANIFESTTRAITVAR=matrix(c(.3,.1,0,.2,  0,.2,0,.15, 
#  0,0,0,0, 0,0,0,.4) ,n.manifest,n.manifest),
#  MANIFESTMEANS=matrix(c(0,0,0,0),n.manifest,1),
#  DRIFT=matrix(c(-.23,.1,.0,-.4),n.latent),
#  DIFFUSION=matrix(c(8.3,5.1,0,8.4),n.latent,n.latent),
#  CINT=matrix(c(0,.4),n.latent,1),
#  TDPREDEFFECT=matrix(c(1.2,-.4),nrow=n.latent,ncol=n.TDpred),
#  TIPREDEFFECT=matrix(c(.32,-.08),nrow=n.latent,ncol=n.TIpred),
#  TDPREDMEANS=matrix(c(0,0,1,rep(0, (Tpoints-1-3)*n.TDpred)),nrow=n.TDpred*(Tpoints-1)),
#  TIPREDMEANS=matrix(0,nrow=n.TIpred),
#  TDPREDVAR=diag(0.1,n.TDpred*(Tpoints-1)),
#  TIPREDVAR=diag(.4,n.TIpred),
#  T0MEANS=matrix(0,ncol=1,nrow=n.latent))
#  
# data<-ctGenerate(generatingModel,n.subjects=100,burnin=500)
# 
# model<-ctModel(n.latent=n.latent, n.TDpred=n.TDpred, n.TIpred=n.TIpred, 
# n.manifest=n.manifest, 
# LAMBDA=matrix(c(1,"l2","l3",0,0,0,0,1),n.manifest,n.latent),  
# TRAITVAR='auto',Tpoints=Tpoints)
# 
# 
# fit<-ctFit(data,model, stationary=c('T0VAR','T0MEANS', 'T0TIPREDEFFECT'))
# summary(checkf)  
# ## End(Not run)