1
$\begingroup$

I runned a fixed effect regression in Stata (xtreg, fe) for panel data. I obtained below R-sq. results. How should I interpret them? Which one is my main interest(within,below or overall)? How can I derive the exact Adjusted R-sq. by using these 3 values?

R-sq:
within = 0.8226
between = 0.4727
overall = 0.5311

Thanks a lot!

$\endgroup$
  • $\begingroup$ In general, you should care most about the within R^2. $\endgroup$ – E. Sommer Feb 28 at 8:19
2
$\begingroup$

In panel regressions you have multiple dimensions and that is why also you have 3 different $R^2$. The within $R^2$ tells you how much variation within your panel variables is on average explained by your model. The between $R^2$, tells you how much variation between your panel variables is explained by the model, and overall $R^2$ gives you the combination of both and tells you how much variation in the whole panel data your model explains.

According to the stata manual you should get adjusted $R^2$ for fixed effects by ereturn list it will be the r2_a scalar. However, there are no simple adjustments for within and between $R^2$, those are most often reported unadjusted.

| improve this answer | |
$\endgroup$
0
$\begingroup$

The ANOVA results indicate 53.11% of dependent variable is explained - which you should check whether it is significant (or insignificant). Moreover, you may ascertain the chi-square with the help of between and within sum of squares. This could help you to test the significance level at which the effect on dependent variable is significant.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.