1-step method in LCA with continous d... PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Shuai Chen posted on Saturday, April 06, 2013 - 4:06 am
Hello,

I want to do 1-step method in LCA with continous distal outcome x.
My model is: x-->c-->y, where c is latent class, y is observed categorical indicator.

I want to predict the mean of x from c, ie, E(x|c). How to obtain this from Mplus results?

Thank you!
Shuai
 Bengt O. Muthen posted on Saturday, April 06, 2013 - 7:29 pm
I think you get the x means in each class in the output - request RESIDUAL.
 Shuai Chen posted on Wednesday, April 10, 2013 - 12:27 am
Thank you!

One more question, how to obtain E(x|c), if the model changes to be: c-->y?
 Bengt O. Muthen posted on Wednesday, April 10, 2013 - 6:11 pm
You say that your model is c->y, but what role does x play in the model?
 Shuai Chen posted on Thursday, April 11, 2013 - 4:35 am
Here x is not in the model. I guess now we can not use 1-step method since x is not in the model, but can only use 3-step method by setting x as AUXILIARY. Am I right?
 Bengt O. Muthen posted on Thursday, April 11, 2013 - 2:40 pm
Right.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: