|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Nested model comparison for ML/bootst... |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
Student posted on Thursday, April 09, 2015 - 11:00 am
|
|
|
I have nested models for which I am using ML estimator with bootstrapping procedures, as they are mediation models. What is the best way to compare the nested models/conduct Chi square difference testing? My understanding is that 1) bootstrapping methods are recommended for mediation models, 2) bootstrapping helps to deal with non-normal outcomes when using the ML estimator, and 3) for continuous non-normal outcomes, the chi-square difference test for the Satorra-Bentler scaled chi-square is recommended. However, the models (using ML estimator/bootstrapping) do not produce a scaling correction factor that can be used to conduct the scaled difference in Chi Sqr test. Any help is appreciated. Thank you! |
|
|
You can run the model a second time with estimator=mlr and without the bootstrap option to get the SB-LRT testing. |
|
Student posted on Thursday, April 09, 2015 - 2:47 pm
|
|
|
Thank you! |
|
Back to top |
|
|