RJM posted on Thursday, September 15, 2005 - 2:14 pm
I would like to specify a model having an induced latent variable, that is, an endogenous latent variable having no indicators and being wholly defined by its predictors. I imagine the code something like as follows, but this is obviously incorrect:
I think you are asking about formative indicators. Following is a description from an earlier post by Bengt:
"In some applications it may be more natural to let a factor be defined as being influenced by indicators, rather than influencing the indicators. An example is SES. Mplus handles this modeling by the following model statements.