Simulation of LCA with distal outcome PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Jietng Zhang posted on Wednesday, July 03, 2013 - 9:50 am
Dr. Muthen,
I am trying some simulation of LPA with binary distal outcome, in reference of the Appendix A in webnote15 V6. I get stuck with two problems and hope if you could help.

1) I cannot generate binary distal outcome with Auxiliary option. Is there any other option available for this? My na´ve thought is to generate continuous distal outcome ranging from 0-1, and recode it as binary by a threshold. But I have no idea how to manipulate different effect sizes of latent class on distal outcome by this way.

2) It seems external analysis is not available for AUXILIARY option. I may use MplusAutomation for analyzing the 500 simulated data. However, if it is a binary distal outcome, I need to use the manual 3-step, which means I have to calculate the classification error 500 times and generate 500 different inp files. Is that right, or any better way than this?

Thank you in advance.
 Bengt O. Muthen posted on Wednesday, July 03, 2013 - 10:40 am
1) If you change the Appendix A input to

Auxiliary = y(DCAT);

and use Generate and Categorical, things work fine. Or, Auxiliary=y(DCON).

If you have problems, send output to
 Jietng Zhang posted on Wednesday, July 03, 2013 - 1:03 pm
Dr. Muthen, thank you, it works!
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message