Hello,

I am writing a thesis paper on how does income inequality (independent variable) affects economic growth (dependent variable). To express "economic growth" I have found data for 2 variables: i) GDP per capita (GDPpc) and ii) GDP per capita growth (GDPpcgr) and I am not sure which one to use in my regression analysis. Regarding GDP per capita, as it is expressed in really high numbers, I took the logarithm of it (lnGDPpc). However, by taking the logarithm of GDP per capita, does it get transformed to GDP per capita growth?

Additionally, as lnGDPpc was not stationary, I had to take the first difference of it (d1_lnGDPpc) to transform it into stationary. What are the implications of doing this?

The variable GDPpcgr was already in stationary form, so I did not have take its first difference.

I have tested both these variables (d1_lnGDPpc and GDPpcgr) in my regressions and I get better (more reasonable) results with d1_lnGDPpc than GDPpcgr (I get really high coefficients with this one).

Should I continue my regression analysis using d1_lnGDPpc? Is the first difference of the logarithm of GDP per capita good for a dependent variable expressing economic growth?

P.S. Sorry if the way of expressing my question is a bit confusing.

Similar questions and discussions