Heteroscedasticity modeling & Type = ... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Tino Nsenene posted on Monday, September 19, 2016 - 12:50 am

Table 1.15 in the new Mplus-book shows how to do a Linear regression with heteroscedasticity modeling. I'd like to do something similar for a type = twolevel analysis. But it seems that the 'constraint' command is not available for type = twolevel. Is there any other syntax option to model heteroscedasticity when using type = twolevel in Mplus? Will 'constraint' for twolevel analysis become available in Mplus 8?

 Bengt O. Muthen posted on Monday, September 19, 2016 - 11:55 am
Q1. No. The closest you can get is to have a multiple-group analysis where groups are based on categories of the variable that influences the heteroscedasticity.

Q2. Probably not in Version 8.0 because the wish list is long.
 Patrick Gaudreau posted on Tuesday, October 23, 2018 - 11:20 am

On page 45 of the book, it is said that a heteroscedasticity model such as (1.64) can be used in situation where the b parameter is expected to have a negative value.

What are the options available in MPLUS when the b parameter is expected to have a positive value?

Best regards,

 Bengt O. Muthen posted on Tuesday, October 23, 2018 - 11:29 am
The same approach can be used.
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