St Error and EM algorithm PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Sanjoy posted on Thursday, December 08, 2005 - 1:01 am
Professor ...If I am not wrong, then we can not calculate SE while doing EM algorithm ... if you can explain us how does Mplus calculate SE while estimating a latent class model (I mean LCA) ....

I am expecting this question in my defense, as I have used Mplus to estimate my model

I have gone through your Biometrics paper (1999). You have mentioned about bootstrap SE at the begining of your paper. I have done simple EM (like in GAUSS and SPlus) without SE ... Precisely ...I am wondering how can I do EM as well as bootstraping simultaneously

is it kind of "imputation" thing ?

Thanks and regards
 bmuthen posted on Thursday, December 08, 2005 - 6:32 am
Read the McLachlan-Peel mixture book (Wiley) - you will see descriptions of getting SEs using the methods by Tom Louis and others.
 Sanjoy posted on Thursday, December 08, 2005 - 11:07 pm
Thank you professor ... I just browsed through it (page 68-70)

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