02-09-2018 08:32 AM
I have two SAS proc nlmixed procedures the difference is that they have different Likelihood functions.
My question is that how can I use Hausman test to examines if there is a significant difference between two models.
My logic is this (May not be helpful)...
"A more efficient model,A against a less efficient (regular) but consistent model. The idea is to make sure that the more efficient model also gives consistent results. Under the null hypothesis of this test, the estimated
coefficients (Bo and B1) from A are consistent only if Bo (regular model) is efficient, while under the alternative hypothesis B1 (A model) is consistent."
02-15-2018 05:56 AM
There are lots of threads that indicate that Hausman test is used in time series. FOr example, see
Moving this post to the EtS community to see if those experts can offer advice.