Attachment | Size |
---|---|
path diagram for model [6] | 114.52 KB |
First post by new user: Can anyone give some advice on an empirical under-identification issue please? I am trying to fit the following model:
resVars <- mxPath( from=mylabels, arrows=2, free=TRUE, values=rep(1,12), labelatVars <- mxPath( from=c("X1","X2"), arrows=2, connect="unique.pairs", free=c(TRUE,FALSE,TRUE), values=c(1,0,1), labels=c("varX1","cov","varX2") ) # factor loadings for X1 facLoadsX1 <- mxPath( from="X1", to=mylabels, arrows=1, free=c(FALSE,TRUE,TRUE,TRUE,TRUE,TRUE,FALSE,TRUE,TRUE,TRUE,TRUE,TRUE), values=rep(1,12), labels =c("l1","l2","l3","l4","l5","l6","l1","l2","l3","l4","l5","l6") ) # factor loadings for X2 facLoadsX2 <- mxPath( from="X2", to=mylabels, arrows=1, free=c(FALSE,TRUE,TRUE,TRUE,TRUE,TRUE,FALSE,TRUE,TRUE,TRUE,TRUE,TRUE), values=rep(1,12), labels =c("k1","k2","k3","k4","k5","k6","k1","k2","k3","k4","k5","k6") ) # means means <- mxPath( from="one", to=c(mylabels,'X1','X2'), arrows=1, free=c(T,T,T,T,T,T,T,T,T,T,T,T,FALSE,FALSE), values=c(1,1,1,1,1,1,1,1,1,1,1,1,0,0), labels =c("meanA","meanB","meanC", "meanD","meanE","meanF","meanA","meanB","meanC", "meanD","meanE","meanF",NA,NA) ) twoFactorModel <- mxModel("Two Factor Model", type="RAM", manifestVars=mylabels, latentVars=c("X1","X2"), dataRaw, resVars, latVars, facLoadsX1, facLoadsX2, means) twoFactorFit <- mxRun(twoFactorModel) ls=c("e1","e2","e3","e4","e5","e6","e1","e2","e3","e4","e5","e6") )
I find that this model is underidentified and has large Std errors which is presumably a result of the underidentification. I think that there should be some additional constraints but would appreciate any advice?