Title: Maddison Project Data
Version: 1.0.2
Date: 2025-11-20
Description: Relatively easy access is provided to Maddison project data, which collates all the credible data on population and GDP for 169 countries, with some dating back to the year 1. 'MaddisonLeaders' makes it easy to find the leaders for each year, allowing users to delete countries like OPEC with narrow economies to focus on the technology leaders. 'ggplotPath' makes it easy to plot data for only selected countries or years.
License: MIT + file LICENSE
URL: https://github.com/sbgraves237/MaddisonData
BugReports: https://github.com/sbgraves237/MaddisonData/issues
Depends: R (≥ 4.1)
Suggests: ggplot2, ipumsr, KFAS, knitr, lubridate, readxl, rmarkdown, testthat (≥ 3.0.0), tibble, usethis
VignetteBuilder: knitr
Config/testthat/edition: 3
Encoding: UTF-8
LazyData: true
RoxygenNote: 7.3.3
NeedsCompilation: no
Packaged: 2025-11-20 13:47:13 UTC; sg
Author: Spencer Graves ORCID iD [aut, cre]
Maintainer: Spencer Graves <spencer.graves@effectivedefense.org>
Repository: CRAN
Date/Publication: 2025-11-24 18:10:08 UTC

Convert a vector of date ranges into a data.frame

Description

MadDateRanges returns a data.frame with 3 numeric columns: yearBegin, yearEnd, and sourceNum from the vector of dateRanges associated with different sources in MaddisonSources.

Usage

MadDateRanges(dateRanges)

Arguments

dateRanges

character vector of date ranges, each associated with a different source.

Value

a data.frame with 3 columns

yearBegin, yearEnd

numeric years

sourceNum

1, 2, 3, ... for the location in dateRanges

Examples

MadDateRanges(c('1', '700 – 1500', '1252–1700 (England)', 
      '1915-1919 & 1949', '1820, 1870, 1913, 1950'))
# equal 
data.frame(
yearBegin=c(1,  700, 1252, 1820, 1870, 1913, 1950), 
yearEnd  =c(1, 1500, 1700, 1820, 1870, 1913, 1950), 
sourceNum=c(1, 2, 3, rep(4, 4)))

Maddison Project data

Description

The Maddison project collates historical economic statistics from many sources. MaddisonCountries is a data.frame of all (countrycode, country, region) combinations in those data.

Usage

MaddisonCountries

Format

MaddisonCountries

A data frame with 3 columns:

ISO

3-letter ISO country code

country

Country name used by the Maddison project

region

Geographic region including country

Its rownames = ISO.

Source

https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2020?lang=en"Groningen Growth and Development Centre"

Examples

# Get the country for a countrycode (IS)
subset(MaddisonCountries, ISO=='GBR', country)
# Or
MaddisonCountries['GBR', 'country']
# Find Yugoslavia 
subset(MaddisonCountries, grepl('Yugo', country), 1:3)
# number of countries by region 
table(MaddisonCountries$region)
# What are "Western Offshoots"? 
subset(MaddisonCountries, grepl('Of', region), c(country, ISO))

Maddison Project data

Description

The Maddison project collates historical economic statistics from many sources. MaddisonCountries is a data.frame of all (countrycode, country, region) combinations in those data.

Usage

MaddisonData

Format

MaddisonData

A data frame with 4 columns:

ISO

3-letter ISO country code

year

numeric year starting with year 1 CE

gdppc

Gross domestic product (GDP) per capita in 2011 dollars at purchasing power parity (PPP)

pop

Population, mid-year (thousands)

Source

https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2020?lang=en"Groningen Growth and Development Centre"

Examples

# Get the countrycode for a country
subset(MaddisonCountries, country=='United Kingdom', ISO)
# Select  
str(GBR <- MaddisonData[MaddisonData$ISO=='GBR', ])

Plot selected countries

Description

MaddisonLeaders computes the countries with the highest gdppc for each year.

Usage

MaddisonLeaders(
  except = character(0),
  y = "gdppc",
  group = "ISO",
  data = MaddisonData::MaddisonData,
  x = "year"
)

Arguments

except

either NULL to select all the data in MaddisonData or a character vector of group codes to EXCLUDE, e.g., so the result reflects apparent technology leaders, excluding countries whose high gdppc may be due to a dominant position in a single commodity.

y

name of column in data to consider. Default = gdppc.

group

name of column in data as the grouping variable. Default = ISO.

data

data.frame or tibble::tibble with first two columns being ISO and year and y being the name of another column.

x

time variable. Default = year.

Value

an data.frame with columns

(defaults:

with an attribute LeaderByYear = a data.frame with columns, {{x}}, paste0('max', y), and {{group}} (defaults: year, maxgdppc, ISO).

Examples

Leaders0 <- MaddisonLeaders() # max GDPpc for each year. 

# Presumed technology leaders without commodity leaders with narrow 
# economies 
Leaders1 <- MaddisonLeaders(c('ARE', 'KWT', 'QAT')) 
# since 1600 
MadDat1600 <- subset(MaddisonData, year>1600)
Leaders1600 <- MaddisonLeaders(c('ARE', 'KWT', 'QAT'), data=MadDat1600)

Maddison Project data

Description

The Maddison project collates historical economic statistics from many sources. MaddisonSources is a list of tibble::tibbles with ISO names giving the sources of GDP per capita for different years for the said country.

MaddisonYears is a data.frame giving yearBegin and yearEnd and the number of each source in MaddisonSpources for each ISO.

Usage

MaddisonSources

MaddisonYears

Format

MaddisonSources

A named list of tibble::tibbles, one for each country, named with the ISO country codes. Each tibble has one row for each source for the indicated ISO and two columns:

years

character variable of year(s) for this source starting with year 1 CE.

source

character variable giving the source for the years described.

In addition, MaddisonSources has an attribute since2008, which says, "gdppc since 2008: Total Economy Database (TED) from the Conference Board for all countries included in TED and UN national accounts statistics for all others."

MaddisonYears

A data.frames with 4 columns:

ISO

3-letter country code.

yearBegin, yearEnd

Integer year begin and end for each source.

sourceNum

Integer of the source within MaddisonSources[[ISO]].

An object of class data.frame with 133 rows and 4 columns.

Source

https://www.rug.nl/ggdc/historicaldevelopment/maddison/releases/maddison-project-database-2020?lang=en"Groningen Growth and Development Centre"

Examples

MaddisonSources[['GBR']]
MaddisonSources[['GBR']][, 1, drop=TRUE] 
# = c('1', '1252–1700 (England)', '1700–1870') 
# for data from the year 1 
# and for England only between 1252 and 1700, etc. 

MaddisonSources[['IRN']][, 1, drop=TRUE] 
# = '1820, 1870, 1913, 1950'
# for those 4 years only. 

MaddisonSources[c('GBR', 'USA')]

MaddisonSources[['GBR']][, 1, drop=TRUE] 
# = c('1', '1252–1700 (England)', '1700–1870') 

MaddisonYears[MaddisonYears$ISO=='GBR', ] = 
data.frame(
ISO=rep('GBR', 3), 
yearBegin=c(1, 1252, 1700), 
yearEnd  =c(1, 1700, 1870), 
sourceNum=1:3
)

MaddisonSources[['EGY']][, 1, drop=TRUE] 
# = c('1', '700 – 1500', '1820, 1870, 1913, 1950')

MaddisonYears[MaddisonYears$ISO=='EGY', ] = 
data.frame(
ISO=rep('EGY', 6), 
yearBegin=c(1,  700, 1820, 1870, 1913, 1950), 
yearEnd  =c(1, 1500, 1820, 1870, 1913, 1950), 
sourceNum=c(1,    2, rep(3, 4))
)

Get Maddison sources

Description

The Maddison project collates historical economic statistics from many sources.

They have a citation policy: CONDITIONS UNDER WHICH ALL ORIGINAL PAPERS MUST BE CITED:

a) If the data is shown in any graphical form b) If subsets of the full dataset that include less than a dozen (12) countries are used for statistical analysis or any other purposes

When neither a) or b) apply, then the MDP as a whole can be cited.

getMaddisonSources returns a data.frame of relevant sources for a particular application.

Usage

getMaddisonSources(
  ISO = NULL,
  plot = TRUE,
  sources = MaddisonData::MaddisonSources,
  years = MaddisonData::MaddisonYears
)

Arguments

ISO

either NULL to return all sources or a character vector of ISO codes for the countries included in the analysis or a data.frame with the first column being the ISO codes followed by yearBegin and optionally yearEnd.

plot

logical indicating whether the use does nor does not include plotting data. The Maddison project requires citing all relevant MaddisonSources if they are plotted, denoted here by plot = TRUE. If no data are plotted, then the Maddison project requires citing all sources only if less than a dozen are used, denoted here by plot = FALSE, in which case, the Maddison project requires a specific project-level citation. Default = TRUE.

sources

list of sources in the format of MaddisonSources; default is MaddisonSources.

years

data.frame in the format of MaddisonYears; default is MaddisonYears.

Value

a tibble::tibble with 3 columns:

ISO

3-letter ISO code for country.

years

character vector of years or year ranges for which source applies.

source

character vector of sources.

in the format of MaddisonSources.

Examples

getMaddisonSources() # all 
getMaddisonSources(plot=FALSE) # only MDP 
getMaddisonSources('GBR') # GBR 
getMaddisonSources(names(MaddisonSources)[1:12], FALSE) # only MDP 
getMaddisonSources(data.frame(ISO=c('GBR', 'USA'), 
             yearBegin=rep(1500, 2)) ) #GBR, USA since 1500 
getMaddisonSources('AUS') # AUS: no special sources for AUS.

ggplot paths

Description

ggplotPath plots y vs. x (typically year) with a separate line for each group with options for legend placement, vertical lines and labels.

Usage

ggplotPath(
  x = "year",
  y,
  group,
  data,
  scaley = 1,
  logy = TRUE,
  legend.position,
  vlines,
  labels
)

Arguments

x

name of column in data to pass as x in aes(x=.data[[x]], ...); default = year.

y

name of column in data to pass as y in aes(y=.data[[y]], ...); must be supplied.

group

name of grouping variable, i.e., plot a separate line for each level of group using aes(group=.data[[group]], ...), unless group is missing or length(unique(data[, group])) = 1.

data

data.frame or tibble::tibble with columns x, y, and group.

scaley

factor to divide y by for plotting. Default = 1, but for data in monetary terms, e.g., for MaddisonData, y = 'gdppc' is Gross domestic product (GDP) per capita in 2011 dollars at purchasing power parity (PPP), for which we typically want scaley = 1000.

logy

logical: if TRUE, y axis is on a log scale; default = TRUE.'

legend.position

argument passed to ggplot2::theme. Default depends on ⁠nGps <- length(unique(data[, group]⁠: If nGps = 1, there is no legend. If nGps > 10, legend.position = 'right'. In between, legend.position = c(.15, .5) = center left. For alternatives, see ggplot2::theme.

vlines

= locations on the x axis for vertical lines using ggplot2::geom_vline(aes(xintercept = .data[[x]]), data=vlines, ...) with ⁠color='grey', lty='dotted'⁠ unless color or colour and / or lty are available as attr(x, ...).

labels

= data.frame with columns ⁠x, y, label, srt, col⁠, where x, y, and srt are numeric, label is character, and col are acceptable values for color in with(labels, annotate('text', x=x, y=y, label = label, srt=srt, color=col)).

Value

an object of class ggplot2::ggplot, which can be subsequently edited, and whose print method produces the desired plot.

Examples

str(GBR_USA <- subset(MaddisonData::MaddisonData, ISO %in% c('GBR', 'USA')))
GBR_USA1 <- ggplotPath('year', 'gdppc', 'ISO', GBR_USA, 1000)

GBR_USA1+ggplot2::coord_cartesian(xlim=c(1500, 1850)) # for only 1500-1850 
GBR_USA1+ggplot2::coord_cartesian(xlim=c(1600, 1700), ylim=c(7, 17)) 

# label the lines
ISOll <- data.frame(x=c(1500, 1750), y=c(1.4, .7), label=c('GBR', 'USA') )
(GBR_USA2 <- ggplotPath('year', 'gdppc', 'ISO', GBR_USA, 1000, 
                        labels=ISOll) ) 
# vlines 

Vlines = c(1849, 1929, 1933, 1939, 1945)
(GBR_USA3 <- ggplotPath('year', 'gdppc', 'ISO', GBR_USA, 1000, 
                vlines=Vlines, labels=ISOll) )

Select countries and add logged variables

Description

logMaddison returns a tibble::tibble of data on selected countries extracted from MaddisonData, appending columns lnGDPpc and lnPop = natural logarithms of gdppc and pop.

Usage

logMaddison(ISO = NULL)

Arguments

ISO

either NULL to select all the data in MaddisonData or a character vector of ISO codes used in the Maddison project.

Value

a tibble::tibble with 6 columns:

ISO

3-letter ISO code for countries selected

year

numeric year in the current era.

gdppc

Gross domestic product per capita adjusted for inflation to 2011 dollars at purchasing power parity.

pop

Population, mid-year (thousands)

lnGDPpc

log(gdppc)

lnPop

log(pop)

Examples

logMaddison() # all 
logMaddison(c('GBR', 'USA')) # GBR, USA

Construct a path to a location within an installed or development package

Description

path_package2 returns a character vector of matches to target. It differs from system.file() in that it supports searching for a target file or folder possibly in subdirs of the working directory or in nparents of its parents.

Usage

path_package2(
  target,
  package = NULL,
  nparents = 1,
  subdirs = c("extdata", paste("inst", "extdata", sep = .Platform$file.sep))
)

Arguments

target

A regular expression describing the file of folder desired.

package

Name of the package to in which to search. If NULL, search in the working directory. Otherwise search in system.file(package).

nparents

integer indicate the number of parents of the working directory in which to search; default = 1.

subdirs

= c('extdata', paste('inst','extdata', sep=.Platform$file.sep))

Details

This works in a vignette searching for a target that could be in the vignettes directory of its parent package or in the package directory or in, e.g., one of subdirs = c('extdata', paste('inst', 'extdata', sep=.Platform$file.sep)).

Returns the full path to match(s) if found and a character vector of length 0 if no matches are found. The returned object also has a searched attribute being a character vector of the directories searched.

This was inspired by a desire to share with others a vignette describing how to create data objects from a file that could not itself be shared on CRAN. This is not easy, because the working director available to code in a vignette changes depending on how that code is run.

path_package2 allows the user to store the target locally, e.g., in inst/extdata but include it in .gitignore to prevent it from leaving the local computer. The vignette then decides what to do after calling path_package2() based on the length of the the object returned.

Value

a character vector with an attribute searched giving the full paths of all directories searched for target.

Examples

# search for a file matching a regular expression
path_package2('^mpd.*xlsx$') 
# search only in the working directory
path_package2('^mpd.*xlsx$', nparents=0, subdirs=character(0))