Page 1 of 1

Latent variables without demographic predictors?

Posted: 08 Feb 2024, 05:34
by Koolau
Greetings. I've estimated hybrid choice models in Mplus and would like to do so in Apollo. I've worked my way up from MNL to ML and now to HCM in Apollo. As the first step, I'm trying to estimate a simple HC model -- similar to the "Hybrid choice model on drug choice data, using continuous measurement model for indicators" example, but without demographic predictors of the latent variable and associated definition of a random coefficient for LV. This was problematic, with the following message: "The estimation of the scaled model failed, and the unscaled version will be returned instead."

Is it possible to estimate HC models in Apollo with latent variables being predicted by other latent variables but not by demographic variables? If so, is there an example available to illustrate the syntax?

Thank you very much for any tips on this.

Re: Latent variables without demographic predictors?

Posted: 05 May 2024, 12:02
by stephanehess
Hi

yes, this is no problem.

It's not likely to be the cause for your issue. Can you show us the estimates?

Stephane