Dear Sriteja

Fixed vs random Hausman test in its standard formulation demands a set of rather strict conditions to be met (you may find them e.g. in Wooldridge “Econometric Analysis of Cross Section and Panel Data”) which is probably the source of your problem. I would suggest to use ‘xtoverid’ – a user written procedure – it should help.

When you save the coefficients and run the test sometimes I've found myself getting an error, I dont remember if is the same as the one you got, but when I've changed the order from which I've call them I get to fix the problem:

xtreg .... , re

est store re

xtreg .... , fe

est store fe

hausman fe re

Hope it helps

Hey,

if you use STATA you can also use the sigmaless or sigmamore option to use the variance covariance matrix of the RE estimates or the FE estimates.

Kind regards,

Florian

I faced the same problem. So i used the following commands.

xtreg .... , fe

estimates store fixed

xtreg .... , re

estimates store random

hausman fixed random,sigmamore

Regards.

You can solve it by using sigmamore after hausman command in STATA. So, it will be "hausman fe re, sigmamore".

hi

I think it can solves by using sigmamore after hausman command in STATA.

So, it will be "hausman fe re, sigmamore".

solved by using sigmamore ,however i am not sure of the difference between sigmamore and sigmaless. both do work, but any idea/reasoning behind this ?

i faced the same problem. I used hausman fe re, sigmamore / sigmaless command but still doesn't work, got the same note. And also hausman fixed random, sigmamore doesn't exist. Is it possible the data are not on equal amount? I attached the data below. Would highly appreciate if you can help me fix this.

simply, you can use Ordinary least squares. There is no need to use fe or re.