Hi all,

In my research i'm running a multilevel model. I used this syntax:

Code:
mixed sqrt_Relative_time_days White Hispanic Other Age Sex i.Seriousness_of_crime_7 Death_Penalty i.year_of_conviction_groups  || states_combined:
Afterwards, I wanted to run the Hausman test in order to compare the coefficients of the multilevel model to the coefficients of an OLS regression.


Code:
regress sqrt_Relative_time_days White Hispanic Other Age Sex i.Seriousness_of_crime_7 Death_Penalty i.year_of_conviction_groups
estimates store fixed

mixed sqrt_Relative_time_days White Hispanic Other Age Sex i.Seriousness_of_crime_7 Death_Penalty i.year_of_conviction_groups  || states_combined:
estimates store random

hausman fixed random, equations(1:1)
However, on the results I got a message saying that -
Code:
model fitted on these data fails to meet the asymptotic assumptions of the Hausman test; see suest for a generalized test".
As I understand, this is a pretty common problem. Since the Hausman test didn't work, I tried to use the suest command:

Code:
suest fixed random
but I got this message:
Code:
mixed is not supported by suest
As you can see, I can't use the Hausman test nor the suest command. Is there anyone familiar with this kind of problem, and knows how can I conduct my estimation after a multilevel model?

Thanks!