ECON 385: Replication Exercise
Fall 2019
Due on Canvas as a pdf: Nov. 19th at 8:00 PM
In this replication exercise you will be replicating one of the most cited papers in Economic growth:
A Contribution to the Empirics of Economic Growth by N. Gregory Mankiw; David Romer ;
David N. Weil. The Quarterly Journal of Economics, Volume 107, Issue 2, 1 May 1992, Pages
407–437.
A. The pdf file for this paper is posted on Canvas. You will need to read the details before each table you
need to replicate. Please ignore the theoretical discussion and only focus on the empirical discussion
before each table. All you need is to figure out how to create dependent and independent variables for
estimating the model using OLS.
B. The data set is provided for three samples on Canvas:
a) NonOil.csv
b) Intermediate.csv
c) OECD.csv
The tables you will be replicating in this exercise have 3 columns. Each column correspond to a specific data
set. I have named the csv files to reflect which sample should be used for replicating a given column of each
table.
C. In each of the these CSV files there is data used by the authors on the following variables:
• CountriD=Country ID number.
• GDP60=Per capita GDP in 1960.
• GDP85=Per capita GDP in 1985.
• POPGRO=Average growth rate of working-age population 1960 to 1985.
• IONY=Average ratio of investment (including Government Investment) to GDP from 1960 to
1985.
• SCHOOL=Average fraction of working-age population enrolled in secondary school from 1960 to
1985.
D. Your task is to replicate (as closely as possible)
• Tables II, III, IV, and V in the pdf file posted on Canvas.
• Figure 1 (all three panels).
1) In your answer, use exactly same format for the table as used in the paper. DO NOT copy-paste your
R output. I will cut points for not providing results in the proper format.
2) Briefly comment on how closely you were able to replicate the beta coefficients for each sample.
3) In the appendix (at the end) provide your complete R code.
GOOD LUCK!