Correlating World Conflict with Local Economy Growth

Overview

For the average student, the Uppsala Conflict Data Program <https://ucdp.uu.se/> offers a fascinating periscope onto the conflicts of the world. Together with economic figures as collected by Statistik Austria <https://www.statistik.at>, the local economic impact of world-wide conflict is calculated.

WARNING: While this student is fairly confident regarding the numerical quality of the calculations, any conclusions have to be regarded as having worse-than-questionable statistical validity.

This project is also covered by an included Data Management Plan (also available as machine actionable version).

Public Project Repository

Find the full content of this project, including content linked in this README, at:

Public Project Repository DOI

https://doi.org/10.5281/zenodo.4699808

Public Project Repository URL

https://github.com/qyanu/2021-gdp-vs-ucdp

How to Build

1. Optionally, reproduce the manual download steps as described in file sna-at-download

   Note: this git repository should already contain the results from a previous manual download of the SNA-AT data.

2. Optionally, reproduce the manual extract & combine steps as described in file sna-at-extract-and-combine

   Note: this git repository should already contain the results from a previous extract & combine process.

3. Execute the build script:

   $ build.sh
   

4. Perform data quality verification according to the UCDP Data Quality Checklist

Build Output Description

directory output-data/ ... the final output and result.

directory ucdp-data ... the UCDP data as downloaded from UCDP webpage.

directory ucdp-data-extracted ... the intermediate data based on the UCDP data.

directory sna-at-data ... the SNA-AT data as downloaded from STATISTIK AUSTRIA webpage.

directory sna-at-data-extracted ... the intermediate data based on the SNA-AT data.

Build Dependencies

• bash5
• python3.7
• dot from the graphviz package v2.40.1
• python3-matplotlib v3.0.2-2
• python3-ruamel.yaml v0.15.34-1+b1
• rst2pdf v0.93.dev0
• curl v7.64.0
• various tools from the coreutils package v8.30 are used.

Note: this project was developed on a Debian Linux "Buster" 10 operating system with a K Desktop Environment.

Development Dependencies

Only needed for scripts other than build.sh, which are not intended to be used by the normal user.

• pip3 install 'jsonschema>=3.2.0'

Intended Applicability

This work is intended to satisfy the curiosity of this student, and does not lend itself to general deliberation. However, one hopes to further the knowledge about the UCDP and suggests to all readers to explore the insightes offered by the UCDP and related resources.

Ingested and Produced Data

Ingested data sets:

• UCDP/PRIO Armed Conflict Dataset version 20.1, from the UCDP at Uppsala University, Sweden
  • Pettersson, Therese & Magnus Öberg (2020) Organized violence, 1989-2019. Journal of Peace Research 57(4).
  • Gleditsch, Nils Petter, Peter Wallensteen, Mikael Eriksson, Margareta Sollenberg, and Håvard Strand (2002) Armed Conflict 1946-2001: A New Dataset. Journal of Peace Research 39(5).
• Volkswirtschaftliche Gesamtrechnungen, from STATISTIK AUSTRIA, Austria

  • Volkswirtschaftliche Gesamtrechnungen 1976-2007 (Standardpublikationen),

    Statistik Austria, Austria

  • Volkswirtschaftliche Gesamtrechnungen 1995 - 2018 (Standardpublikationen),

    from Statistik Austria, Austria

  • Volkswirtschaftliche Gesamtrechnung Hauptgrößen, from Statistik Austria,

    Austria; as published on 2021-04-14

Note, that the ODATA-portal operated by Statistik Austria does offer many data sets via a machine-readable API, however the sought after GDP data set is not available there.

The data from Statistik Austria is filtered and combined as follows:

• from "Volkswirtschaftliche Gesamtrechnungen 1976-2007" the data for the years 1976 to 1994 are used.
• from "Volkswirtschaftliche Gesamtrechnungen 1995 - 2018" the data for the years 1995 to 2015 are used.
• from "Volkswirtschaftliche Gesamtrechnung Hauptgrößen" the data for years 2016 to 2019 are used.
• all three data are combined, in order, into a single table, and the yearly gdp increase is calculated. see file sna-at-data-combined/gdp-at-data.csv. (See file extract-and-combine-sna-at.pdf for details.)

From the UCDP the sum of all intensity level of all conflicts active each year is used.

Produced Data

Cross calculates a x-y plot, comparing both input data sets for direct calculation. This plot is generated as machine-parseable csv file, and as human displayable svg file. See directory output-data/.

Results

Raw scatter plot points: output-data/output-data.csv

There is no discernable correlation of yearly gdp change and yearly world conflict intensity.