Pesaran cross-sectional test
Posted: Sun Mar 27, 2016 11:22 am
Hello,
I woulld like to proceed this in Eviews 8, could you please tell me step by step how to do that:
"Applying standard Dickey-Fuller test to panels brings bias which is not present
in univariate series. There are two classes of panel unit-root tests, rst and
second generation unit root tests, which dier in the assumption of the crosssectional
independence. In the rst generation, there is imposed restriction
that all cross-sections are independent, while second generation of panel unitroot
tests relaxes this assumption. First generation tests are biased if they
are applied to panels with cross-sectional dependency.(Breitung and Pesaran,
2005)
We implement Pesaran cross-sectional dependence (CADF) test to our dataset
to decide whether we should use rst or second generation panel unit-root tests.
The null hypothesis of cross-sectional independence is strongly rejected for all
our time series at 1% signicance level." (Halova, 2015)
Is there any way how to test the cross-sectional dependence without the equation or model set?
Thank you very much for answer.
Romana
I woulld like to proceed this in Eviews 8, could you please tell me step by step how to do that:
"Applying standard Dickey-Fuller test to panels brings bias which is not present
in univariate series. There are two classes of panel unit-root tests, rst and
second generation unit root tests, which dier in the assumption of the crosssectional
independence. In the rst generation, there is imposed restriction
that all cross-sections are independent, while second generation of panel unitroot
tests relaxes this assumption. First generation tests are biased if they
are applied to panels with cross-sectional dependency.(Breitung and Pesaran,
2005)
We implement Pesaran cross-sectional dependence (CADF) test to our dataset
to decide whether we should use rst or second generation panel unit-root tests.
The null hypothesis of cross-sectional independence is strongly rejected for all
our time series at 1% signicance level." (Halova, 2015)
Is there any way how to test the cross-sectional dependence without the equation or model set?
Thank you very much for answer.
Romana