problems with panel data regression

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

pokersellers
Posts: 1
Joined: Thu Nov 13, 2008 4:01 am

problems with panel data regression

Postby pokersellers » Thu Nov 13, 2008 5:59 am

Dear forum users,

I am a student from Hamburg, Germany and currently working on my master thesis, which deals
with a panel data analysis/regression of economic data.
Dependent variable is the container throughput in Hamburg, other variables are GDP, Trade balance, Export etc...

Unfortunately some problems arised, I do not know how to deal with.

To carry out my analysis I opened a new workfile (dated - regular frequency) and inside this workfile
I created a new pool-object with a couple of Cross Section Identifiers (depending on the number of
countries involved) and additionally I introduced a list of ordinary and pool series (spreadsheet, stacked data), where I inserted all necessary data.
To perform the calculations I pressed the button estimate in the pool-schema and tested my model under different conditions (FE, RE, etc)

The results appearing are quite 'nice'. They have a high R-squared and most of the coefficients are highly
significant.

But I do not know how to make other tests (esp. for Heteroskedascity and auto-correlation), because the options given in the pool-'View'-menu differ from
the ones given in the normal 'estimate equation'-section.


Eg there is no White-test.

Does anybody of you know how to perform this calculation correctly?

Thanks and kind regards,

Adam

Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 24 guests