Adjusted R-square is negative?

[Raymond Zhang]

Dear all,
When I run panel data fixed effects model,I get a negative adjusted R2.I don't know why, and is it valid? How can I avoid negative adjusted R2?
Thank you in advance.

Best regards.
Raymond

[Carlo Lazzaro]

Raymond:
the R2adj formula reported in -regress- entry, Stata .pdf manual admits negative values (unlike Rsq).
So, nothing sinister there.
That said, I would double-check your model specification.

[David Radwin]

Also see this earlier post: https://www.statalist.org/forums/for...uared-negative

[Raymond Zhang]

Thank you for your help.@Carlo Lazzaro@David Radwin