Make it easier for humans to access data from the Maddison Data
Project in R. Later releases may include vignettes, etc., documenting
analyses using the [KFAS] (Kalman filtering and smoothing,
aka state space) techniques with these data.
Objectives: Make it relatively easy in R to do the following:
Find the countries with the highest gdppc for each
year for which data are available.
Refine “1” by deleting companies with high gdppc
based on something narrow like a commodity, e.g., oil.
Plot the data available on gdppc and / or pop for a
selection of countries, e.g., world leaders.
LATER:
Build a state space / Kalman models for gdppc and
pop for each country in the Maddison project data.
Use Kalman smooth to interpolate and extrapolate (forward but not
backwards) gdppc and pop for each country for
all years that appear anywhere in the Maddison project data.
Identify the world leader in gdppc for each year,
refining “1” using KFAS interpolation.
Identify the world technology leader for each year by evaluating
the gdppc leader for each year and replacing any whose
leadership was narrow like members of OPEC with a country with a
broad-based economy like the US.
You can install the development version of MaddisonData from GitHub with:
# install.packages("pak")
pak::pak("sbgraves237/MaddisonData")[Coming soon.]
library(MaddisonData)
## basic example code