|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Correlation matrix with single indica... |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
Dear all, I am trying to generate a correlation matrix for a publication. My model has latent and single indicator variables. Doing the SEM with MPlus works. Now I am trying to use TECH4 to generate the correlation matrix. This works as long as there are only latent variables. Now I tried to fake a latent variable from a single indicator variable and it tells me NO CONVERGENCE. Is there a way to get a correlation matrix with combined latent and single indicator variables? Thanks! Richard |
|
|
It works to create a latent variable behind a single indicator - just say: f by y; y@0; This puts the y variance into f. |
|
|
That worked. Thanks a lot! |
|
Back to top |
|
|