Calculation R2 after estimation xtgls

[Ahmad shakiba]

Hello, my model is generalized least squares method, but the R2 has not been calculated. How is it calculated?

xtgls arlag c.ibvol##i.spec auto lev lnat btm tenure opinion los matweak,panels(heteroskedastic)

Cross-sectional time-series FGLS regression

Coefficients: generalized least squares
Panels: heteroskedastic
Correlation: no autocorrelation

Estimated covariances = 133 Number of obs = 1,064
Estimated autocorrelations = 0 Number of groups = 133
Estimated coefficients = 12 Time periods = 8
Wald chi2(11) = 682.79
Prob > chi2 = 0.0000

thankyou

[Carlo Lazzaro]

Ahmad:
welcometothis forum.
-xtgls- returns -chi2-.
Just type:

ereturn list

for further details.

[Ahmad shakiba]

Thank you mr lazzaro but how interpret goodnees of fit GLS estimation?

[Carlo Lazzaro]

Ahmad:
see -Waldi chi2- outcome.
That said, if you have a N>T panel dataset, why going -xtgls- instead of -xtreg-?

[Ahmad shakiba]

160 / 5,000 Translation results

My model has hetroskedastic and That's why I estimated my model to GLS by command xtgls . Is there another way to correct report the coefficient of determination (r2) to me? Thank you

[Carlo Lazzaro]

Ahmad:
-xtreg- with -vce(cluster clusterid) option accounts for heteroskedasticity and/or autocorrelation.

[Ahmad shakiba]

Thank you. My model is fixed effect. Do not need a Voldridge test or modified wald ?

[Carlo Lazzaro]

Ahmad:
if you've already detected heteroskedasticity, there's no need for another test.