Economics 612/613
Spring, 2001
David Mitch

			Problem Set 3

1.For the MRW data set, regress the ratio of  ln( GDPA85/GDPA60) on ln(GDPA60).
Do this for all the countries in the sample and for non-oil countries, intermediate, and 
OECD countries respectively.
2.Plot  the ratio of ln(GDPA85/GDPA60) on ln(GDPA60) for the four groups of 
countries specified in question 1.
3.Do the six regressions for the variables listed in Tables IV and V of MRW. Report
the estimated coefficients, beta coefficients, and t-statistics for the variables in  your 
regressions. Plot the residuals from these regressions against ln(gdpa60) as in Figure I B 
and C of MRW (on p.427).
4. Calculate the implied value of  lambda (the convergence rate as defined in equation 13 
on p.422 of MRW) for the regressions you did in questions 1 and 2.
5.To the regressions in question 3 add as an independent variable the interaction between
ln(GDPA60) and  ln(School). 
6.Generate regional dummy variables for countries in each of the following regions:
Africa, Central and South America, Asia, and Europe.  Redo the regressions for questions 
1, 3, and 4 for each of these four regions  (note  that  rather than doing the regressions for 
the n, I, and o group of countries, as in questions 3 and 4 above you should do the 
regressions for each of the four regions specified). 
7.Calculate the implied value of lambda and a confidence interval for it from your results 
in question 5. Based on your calculations of lambda here and from question 4, discuss to 
what extent the OECD countries and the four regions specified in question 6 each 
constitute convergence clubs. 
8.Discuss to what extent the entire group of non-oil countries constitutes a convergence 
club, based on your results from questions 1,2, and 4.