Controlling for country specific effects.

[jotec90]

Hi, I am performing a multiple regression with unbalanced panel data from 70 countries (1960-2009). I am investigating whether certain financial development indicators have an effect on economic growth (GDP per capita), but I was wondering how I control for specific country effects as I am sure this is hindering my results somewhat. I think this should be quite easy and have consulted my econometrics book to no avail. I am using eviews 6 and any help will be greatly welcomed. Thanks in advance.

[jotec90]

I have found out that I need to use dummies so I have to set UK=1 and 0 otherwise, then France=1 and 0 otherwise, but because I have so many countries that is alot of dummies and I am entering them on my excel spreadsheet and then importing them into eViews. My next question is how do I put them all into one regression so I am not just including a dummy variable for each country in each table? My Teacher wants them in one regression and one table, and I just put a note underneath stating that dummies for each country were used? Thanks in advance

[EViews Glenn]

Look at the discussion of panel estimation and fixed effects.

[jotec90]

Thanks, I have had a look but it doesn't really help me. My equation is gdppc c (control variables) (financial development variables) and then where I have imported my data into eViews after controlling for country specific effects I have another 70 odd variables to add and I get near singular matrix error when I try to add them after my other variables. I think I am doing this wrong, can someone help? Thanks in advance