03-08-2017 01:54 PM - edited 03-08-2017 03:44 PM
I need to do Chow test to see whether to run one single overall model vs multiple models for the subsets of observations on EG6.1. PROC AUTOREG is not installed in my enterprise guide, so I think what I can do is to run separate models for all the subsets and then calculate Chow statistics. Does anyone happen to have this experience and know how to code for calculating Chow statistics?
Thanks as always,
03-08-2017 03:45 PM
I saw an example of doing the Chow test here http://support.sas.com/documentation/cdl/en/etsug/63939/HTML/default/viewer.htm#etsug_model_sect052.....
So it is possible to run the model with the overall sample and do the Chow test with breakpoints. But I am not sure about the right way to put breakpoints…