Negative statistic chi2 after hausman test

[Samiratou Kora Chabi]

Hello everyone,
I have panel data and my dependent variable is a dummy variable. N=801 (individuals) and T=8 (periods)
I am running xtlogit regression in Stata. In order to choose between the fixed effect model and random effect model, I run Hausman test, but I got a negative chi2 statistic. Someone can help me?

Thanks

[Carlo Lazzaro]

Semiratou:
welcome to this forum.
Does adding the -sigmaless- or -sigmamore- options change anything?