Comparing Linear and Count Regression PreviousNext
Mplus Discussion > Categorical Data Modeling >
 Aidan G. Wright posted on Sunday, March 25, 2012 - 5:26 pm
Is it appropriate to use the AIC and/or BIC provided by mplus to compare linear regression models (i.e., variables assumed to be continuous normal) to count based regression models estimated using ML in mplus?

If not, why not?

Thank you for your help.
 Linda K. Muthen posted on Sunday, March 25, 2012 - 5:41 pm
No, you should not do this because the loglikelihoods are in different metrics.
 Aidan G. Wright posted on Sunday, March 25, 2012 - 5:58 pm
Thanks for the quick reply. As a follow up, is this true only of mplus? And do you have a reference for a discussion of this? Thank you.
 Linda K. Muthen posted on Monday, March 26, 2012 - 2:06 pm
No, this is not specific to Mplus. This is true in general. See books that contain linear and count regressions.
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