Panel regression with random effects - error "no observations" r(2000)

[Sven Sadlo]

Hi,

I'm working since 2 days on my econometrics assignment and don't know how to solve the following issue.
I have a fairly simple panel data set (artificial data for practice purposes): 100 entities, each observed at 30 time periods. I am asked to run POLS, FE- and RE-regressions.

I used

xtset

to declare the data as panel and the fixed effect regression using

xtreg Y X, fe

worked perfectly fine. However, if I try to run the random effect regression with

xtreg Y X, re

I receive the following error:

. xtreg Y X, re
no observations
r(2000);

I have no idea about the cause of this problem. The panel ID variables are no strings and that the fixed effect regression works confuses me even more.

Finally, a comment on the data set: It is in some sense a strange dataset because for most entities, there is perfect linear correlation (i.e. R² = 1) over time. However, I'm quite sure that this is not the reason for the problem since a friend of mine estimated the random effects model based on this data in Python.

I appreciate your help! Thanks in advance
Sven