|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Residual Covariances (Cate/Cont) |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
Is there any way to allow for local dependence in LPA/LCA with both, continuous and categorical indicators? From what I've gathered there is parameterization=rescov; to allow for residual covariances between categorical indicators and the normal "with" command to allow for covariances between continuous indicators. Trying to allow for the covariance between a continuous and a categorical indicator returns a fatal error. Is there any working alternative in Mplus to allow for these residuals as well? |
|
|
You have to put a factor behind those 2 indicators like we do in UG ex7.16. |
|
Back to top |
|
|