Conversion of NUTS codes

Introduction

This repository contains the source for a data descriptor article on conversion weights for European regional NUTS data (Nomenclature of Territorial Units for Statistics), targeting submission to Nature Scientific Data. The weights make it possible to convert statistics between different NUTS versions and across the NUTS 1/2/3 levels, so that regional time series remain comparable despite the periodic boundary revisions of the classification.

The conversion weights are distributed through the nuts R package, available on rOpenSci: https://github.com/ropensci/nuts. This repository holds the article itself (Quarto source and rendered output) together with the R scripts that generate every figure and data summary it contains.

Repository content

• nuts_article.qmd — Quarto source file for the data descriptor (submit this one).
• nuts_article.pdf — Rendered data descriptor.
• bibliography.bib — References in BibTeX format.
• codes/ — R scripts that generate the figures and data summaries.
• data/ — Input data and shapefiles, generated CSV summaries, and the data/stats/ files that feed the in-text numbers of the manuscript.
• figs/ — Figure outputs plus source files (SVG, MMD).

Rendering the article

# Render to PDF (main output)
quarto render nuts_article.qmd --to scientific-data-pdf

# Render to Word
quarto render nuts_article.qmd --to docx

Running the code

All figure- and data-generating scripts live in codes/ and are driven by a master script, codes/run_all.R. Run it from the project root so that the relative paths (figs/, data/) resolve correctly. Outputs are written to figs/ and data/.

# Run every script
Rscript codes/run_all.R

# Run only scripts that produce a figure
Rscript codes/run_all.R figures

# Run only the validation scripts
Rscript codes/run_all.R validation

# Run one or more named scripts (with or without the .R extension)
Rscript codes/run_all.R nuts_revisions_map validation_eurostat

Inside an interactive R session you can instead source the master script and call run_all():

source("codes/run_all.R")   # defines run_all(); does not auto-run
run_all()                   # run everything
run_all("validation_eurostat")  # run a subset

Individual scripts can also be run on their own from the project root:

Script	Output
codes/country_coverage.R	data/country_coverage_summary.csv, data/country_coverage_matrix.csv
codes/nuts_revisions_map.R	figs/european_nuts_changes.png
codes/robustness_slivers_sim.R	figs/robustness_slivers_sim.png, data/stats/slivers_stats.rds
codes/validation_eurostat.R	figs/validation_eurostat.png, data/stats/roundtrip_stats.rds
codes/validation_gdp_pop.R	figs/validation_gdp_pop_multiyear.png, figs/validation_gdp_pop_singleyear.png, data/stats/gdp_pop_stats.{csv,rds}
codes/intext_numbers.R	data/stats/intext_numbers.rds

Rscript codes/validation_eurostat.R