Panel Data "near singular matrix"

[ADNL]

Hi,

I have a question regarding excel formation of the data. I want a number of countries with data for 1990 to 2010. I want to regress the average growth of GDP per capita form 1990 - 2000 with GDP per capita in 1990 ( to measure the effect of GDP per capita in 1990 on the average growth of 1990-2000). However, it has to be a panel data, so I can implement fixed effects. How can I best do this? The way I have tried is entering data on the row for 1990 and then leaving the remaining years empty (NA). But EViews then gives a "near singular matrix". Using data for all years 1990-2000 and then for 2000-2010, then how I can I measure the effect of 1990 only?

I look forward to your reply. Thank you in advance.

[EViews Gareth]

http://forums.eviews.com/viewtopic.php?f=7&t=74

[ADNL]

Thanks for the link! It is very helpful to get a panel datasheet in eviews. However, I am still not sure how to do a regression regarding the the average of GDP between 1990-2000 and the influence of the GDP per capita data in 1990 on this average. What technique is the best to use in this case? And how to implement the data then?


Thanks in advance.