Dear all,
I’m working on the bivariate ACE model from 2016 Boulder workshop from Hermione Maes ‘twoSATja’ and ‘twoACEja’ (http://ibg.colorado.edu/cdrom2016/maes/UnivariateAnalysis/scriptsOpenMx.shtml). I use questionnaire data and some participants filled this out one or two years later than their twin siblings, this means that some siblings differ in age. How do I adjust the model to allow different ages between twins? I keep running into problems with non-conforming arrays…
Below the parts of the script that I have trouble adapting:
# Select Covariates for Analysis ordData[,'age'] <- ordData[,'age']/100 ordData <- ordData[-which(is.na(ordData$age)),] covVars <- c('age') nc <- 2 # number of covariates
I adapted the above code to:
twindata$age1<-twindata$age1/100 twindata$age2<-twindata$age2/100 covVars <-c('age1', 'age2') nc <-2
But I keep having trouble adapting the defAge, pathB and expMean in code below:
# PREPARE MODEL # ACE Model # Create Matrices for Covariates and linear Regression Coefficients defAge <- mxMatrix( type="Full", nrow=1, ncol=1, free=FALSE, labels=c("data.age"), name="Age" ) pathB <- mxMatrix( type="Full", nrow=nc, ncol=1, free=TRUE, values=.01, label=labBe, name="b" ) # Create Algebra for expected Mean Matrices meanG <- mxMatrix( type="Full", nrow=1, ncol=ntv, free=frMV, values=svMe, labels=labMe, name="meanG" ) expMean <- mxAlgebra( expression= meanG + cbind(t(b%*%Age),t(b%*%Age)), name="expMean" ) thinG <- mxMatrix( type="Full", nrow=nth, ncol=ntvo, free=TRUE, values=svTh, lbound=lbTh, labels=labTh, name="thinG" ) inc <- mxMatrix( type="Lower", nrow=nth, ncol=nth, free=FALSE, values=1, name="inc" ) threG <- mxAlgebra( expression= inc %*% thinG, name="threG"
Thanks for you help!
Printer-friendly version
Try this: