File: mrw_qr.inp

package info (click to toggle)
gretl 2022c-1
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 59,552 kB
  • sloc: ansic: 409,074; sh: 4,449; makefile: 3,222; cpp: 2,777; xml: 599; perl: 364
file content (28 lines) | stat: -rw-r--r-- 868 bytes parent folder | download | duplicates (4)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
# Replicate Table V in Mankiw, Romer and Weil, QJE 1992,
# plus the quantile regressions in R. Ram, "Parametric
# variability in cross-country growth regressions: An
# application of quantile-regression methodology",
# Economics Letters 99 (2008) pp. 387-9.
open mrw.gdt
series ly60 = log(gdp60)
series dlny = log(gdp85) - ly60
series ngd = 0.05 + (popgrow/100)
series lngd = log(ngd)
series linv = log(inv/100)
series lschool = log(school/100)
# set sample to non-oil producing countries
smpl nonoil --dummy
# OLS, as per M, R & W
ols dlny const ly60 linv lngd lschool

# Top quartile
quantreg 0.75 dlny const ly60 linv lngd lschool
# Bottom quartile
quantreg 0.25 dlny const ly60 linv lngd lschool

# Top quartile (robust)
quantreg 0.75 dlny const ly60 linv lngd lschool --robust
# Bottom quartile (robust)
quantreg 0.25 dlny const ly60 linv lngd lschool --robust