|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
How to get the intercept for a latent... |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
Dear Bengt and Linda: I'm estimating a structural model with latent endogenous variables, and I want to get the intercepts for the latent variables. I have tried asking for them like this: [posfunct]; But I have gotten an error message that the model is not identified. I tried to set one of the indicator intercepts to zero as a way of identifying the model, but then it wouldn't converge!! What am I doing wrong here? Thanks very much, and all the best to both of you. Seth Schwartz |
|
|
In a cross-sectional model, means and intercepts of the latent variables cannot be estimated. They are fixed at zero. Only with multiple groups or multiple time points can they be estimated. |
|
Back to top |
|
|