Panel regression OLS and FE

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

Ananym
Posts: 1
Joined: Tue Feb 14, 2017 4:25 am

Panel regression OLS and FE

Postby Ananym » Thu May 25, 2017 2:25 am

Hello!

Currently I am working on my master thesis. Its on the development of inequality and how it is influenced by globalization factors. I created a dataset with explanatory variables (Trade openness, High tech. exp, FDI, market capitalization) and a few controlvariables. My dependent variable ist the log of gini (gini_ln).
I controlled for unit roots by using the fst difference for the relevant variable. Additionally I added two time dummies to controll for cycle effects. And this is how my basic model looks like:

gini_ln c trade(-1) high_tech_exp(-1) fdi market_capitalization gov_exp____of_total_gdp_ r_e labor_adv_edu labor_basic_edu emp_agr(-1) emp_ind(-1) emp_ser(-1) dummy dummy3

Now my problem: With basic OLS everything is fine. The variables react as predicted and as is documented by other research. But the results of the FE model don't look anything like as in other papers. I just don't know what I am doing wrong, since I am unsing an almost equal dataset as other researchers. So I was hoping that one you may know where I made a mistake? (I added the outputs as attachments below)

Thanks !
Attachments
FE.docx
(18.41 KiB) Downloaded 221 times
OLS.docx
(17.83 KiB) Downloaded 218 times

Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 18 guests