|
|
Correlation predicting latent distal |
|
Message/Author |
|
|
Hello. I am trying to run a model which uses the correlation between 2 latent categorical variables to predict a latent categorical distal outcome. Ideally, I'd like to examine the direct effects of both c1 and c2 on the distal (c3) as well as the effect of the correlation of c1 & c2 on the distal. My Overall model reads MODEL: %OVERALL% c1#1 WITH c2#1; c3#1 ON c1#1; c3#1 ON c2#1; The error message I receive is: *** ERROR in Model command. This model is not supported by LOGIT parameterization. Use LOGLINEAR parameterization. When I specify the loglinear parameterization, I get the following erro message: *** ERROR in Model command. This model is not supported by LOGLINEAR parameterization. Use LOGIT parameterization. How might I go about specifying this model? I greatly appreciate your assistance. |
|
|
It sounds like you need to send your input, data, output, and license number to support@statmodel.com. |
|
mpduser1 posted on Thursday, September 24, 2009 - 9:50 am
|
|
|
Any chance you'll be extending the ADDFREQUENCY feature to allow it to be used for sampling zeros with 3-way LOGLINEAR tables (etc.)? Thanks. |
|
|
We can add that to our suggestion list. |
|
Back to top |
|
|