Hello everyone,
Sorry to bother you all.
I am currently running my panel data model with N=37 and T=5. I choose Fixed Effect Model as suggested by Chow and Hausmann Test. All going well, until i did test for heteroskedasticity and autocorrelation. It turns out my data have both problem. As sugested in Hoechle (2007), exactly in conclusion, i understand it is important to test the cross-sectionally dependent to make sure that statistical inherence is valid. So I look up De Hoyos and Sarafidis (2006), from there I learn in order to test cross-sectionally dependent you can use three differente procedures which are Friedman, Frees, and Pesaran.
When I tried the Pesaran, i am kind of confused with the result as you can see below
. xtcsd, pesaran abs
Pesaran's test of cross sectional independence = -0.389, Pr = 1.3030
Average absolute value of the off-diagonal elements = 0.445
How I interprete this result ?
On other notes, I tried to using Frees and I got the results like these below :
xtcsd, frees
Frees' test of cross sectional independence = 0.812
|--------------------------------------------------------|
Critical values from Frees' Q distribution
alpha = 0.10 : 0.4892
alpha = 0.05 : 0.6860
alpha = 0.01 : 1.1046
how do interprete this result ? Thank you very much for your help.
Sorry to bother you all.
I am currently running my panel data model with N=37 and T=5. I choose Fixed Effect Model as suggested by Chow and Hausmann Test. All going well, until i did test for heteroskedasticity and autocorrelation. It turns out my data have both problem. As sugested in Hoechle (2007), exactly in conclusion, i understand it is important to test the cross-sectionally dependent to make sure that statistical inherence is valid. So I look up De Hoyos and Sarafidis (2006), from there I learn in order to test cross-sectionally dependent you can use three differente procedures which are Friedman, Frees, and Pesaran.
When I tried the Pesaran, i am kind of confused with the result as you can see below
. xtcsd, pesaran abs
Pesaran's test of cross sectional independence = -0.389, Pr = 1.3030
Average absolute value of the off-diagonal elements = 0.445
How I interprete this result ?
On other notes, I tried to using Frees and I got the results like these below :
xtcsd, frees
Frees' test of cross sectional independence = 0.812
|--------------------------------------------------------|
Critical values from Frees' Q distribution
alpha = 0.10 : 0.4892
alpha = 0.05 : 0.6860
alpha = 0.01 : 1.1046
how do interprete this result ? Thank you very much for your help.
Comment