EVIEWS REGRESSION PANEL DATA

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

Angeloti
Posts: 1
Joined: Fri Apr 17, 2020 3:03 am

EVIEWS REGRESSION PANEL DATA

Postby Angeloti » Fri Apr 17, 2020 9:14 am

Hello everyone,
I am writing a dissertation about Capital Structure and determinants of differentiation across industries.
Panel data: 6.000 companies, 5 years (2013-2017), 3 depended viariables(leverage, long leverage and short leverage), 6 independed variables and three dummies as I have data for 4 sectors.
I am a new user in Eviews and I face two difficulties.
1. The first problem is that I made an import all data from an excel file and I'm not sure if eviews recognizes the dummies as dummies and not as values (between 0 to 1) which come from independed variables. I tried to generate dummies in eviews but in panel data form the sample has to do only with the years and not with rows. Specifically, I have 5 years,1500 companies for each sector, so the first dummy which refers to first sector exists from row 1 to 7500, the second dummy from 7501 to 15.000 etc.
2. The second issue is that when i tried to run a regression in cross section fixed model effect the results were significantly better and look artificial to me without including dummies but when I put them in the estimated equation model a message "near singular matrix appeared.

Finally I have been advised to use:
Method: EGLS
GLS weights: Cross-section weights
coef cov method: White cross-section
Can someone please suggest if this is the right way to improve my regression?

Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 21 guests