xtreg result shows no R-square value and f stat value

[UA Munira]

Hello all,

I am trying to run the fixed effect regression for my research but regression result shows no r2 values and f-stat value. I don't understand why this is happening. Can anyone please tell me why this might happen?

Thanks in advance,

Umme

[Clyde Schechter]

Well, you are in the unusual situation that you have a single regressor, and its coefficient is precisely 0. The between and overall R² statistics in Stata are calculated as the squared correlation between the predicted y values and the observed of y, at the group mean or individual observation levels respectively. But in this case, with the sole regression coefficient being zero, there is no variation in the predicted values. The way Stata calculates correlations, the correlation is undefined when one of the variables involved is a constant. Hence the missing R² values. The documentation does not give an explicit formula for the F statistic: there are several ways to calculate it from the regression results. One way involves calculating it from R², and if that is what they are using, it will, therefore, also be undefined.