Downloads

Current and historic forecasts

Due to the size of the data, the complete ViEWS forecasts are currently not available for download. A selection of forecasts are however available through the ViEWS API, our interactive forecasting dashboard, and the resource page on our dedicated website viewsforecasting.org. Limited historic datasets and supplementary material to ViEWS publications are also available for download below.

For access to other data from the ViEWS system, please submit a request to views@pcr.uu.se in which you specify which monthly runs (versions/data releases) that you are interested in, and whether you seek only ensemble results or also constituent model results. The data can be provided in .csv format and/or visualised.

Forecast archives, Jun 2018-Mar 2019

The forecast archives contain forecasts as well as underlying data, as they looked at that point in time. All files are zipped, and contain the data and predictions in separate folders. Note, uncompressed file-sizes are between 2 and 12 GB. The March 2019 run contains data on model evaluation as well as raw data and forecasts. Archives have the format described in this readme file (pdf). The codebooks in the sections below apply also to the forecast archives, unless otherwise stated in the archive itself.

March 2019 (run r19.0.3)(zip)

February 2019 (run r19.0.2)(zip)

January 2019 (run r19.0.1) (zip)

December 2018 (run r18.12.1) (zip)

November 2018 (run r18.11.1) (zip)

October 2018 (run r18.10.1 - This is the run on which the 2020 JPR article is based) (zip)

September 2018 (run r18.9.1) (zip)

August 2018 (run r18.8.1) (zip)

July 2018 (run r18.7.1) (zip)

June 2018 (run r18.6.1) (zip)

Code

The codebase for the ViEWS model is available via the following GitHub repositories:

First version of ViEWS (OpenViEWS): https://github.com/UppsalaConflictDataProgram/OpenViEW S (no longer updated)
Second version of ViEWS (OpenViEWS2): https://github.com/UppsalaConflictDataProgram/OpenV iEWS2 (no longer updated)
Third version of ViEWS:
- FCDO_predicting_fatalities (the fatalities001 model, no longer updated): https://github.com/prio-data/FCDO_predicting_fatalities
- viewsforecasting (the fatalities002 model, and later versions): https://github.com/prio-data/viewsforecasting

Supplementary material

Supplementary material for Hegre et al (2020), Journal of Peace Research 56(2)

Supplementary material for the "ViEWS: A political violence early-warning system”, Journal of Peace Research 56(2).

Database dumps, Oct 2018 and Feb 2020

Full database dumps corresponding to the state of the ViEWS system at the time of the two JPR articles, October 2018 and February 2020:

dbdump-2018-10-20.dmp

dbdump-2020-02-03.dmp

JPR 2019(2) material

Supplementary material for the article "ViEWS: A political violence early-warning system”, Journal of Peace Research 56(2)

Online appendix (pdf)
Forecasts and r eplication data (zip)
Full database dump corresponding to the state of the ViEWS system at the moment of the October 2018 run, the run presented in the JPR 2019(2) article: dbump-2020-02-03.dmp

Civil Society and Democratic Peace

Supplementary material for: Hegre, Håvard, Michael Bernhard & Jan Teorell, 2020. “Civil Society and the Democratic Peace”. Journal of Conflict Resolution, 64(1).

ViEWS outcomes baseline (monthly) at the PRIO-GRID level, 1989-2018

The dataset covers January 1989 to December 2018, and has a one month resolution. Data for 2018 originates from "candidate events".

A description of the data is available in the presentation article.
The ViEWS Outcomes Baseline Monthly codebook (pdf)
ViEWS Monthly Outcomes at PRIO-Grid Level (.zip)

Please cite: Hegre, Håvard, Mihai Croicu, Kristine Eck, and Stina Högbladh, 2018. "Introducing the UCDP-Candidate Events Dataset and the ViEWS Outcomes dataset. Monthly updated organized violence data in the form of events data as well as aggregated to the country-month and PRIO-GRID-month level". Typescript Uppsala University.

ViEWS outcomes baseline (monthly) aggregated at the country level, 1989-2018

The dataset covers January 1989 to December 2018 and has a one month resolution. Data for 2018 originates from "candidate events".

A description of the data is available in the presentation article.
The ViEWS Outcomes Baseline Monthly codebook (pdf)
ViEWS Monthly Outcomes at Country Level (.zip) (currently unavailable, please contact the ViEWS team for more information).

ViEWS outcomes imputed (monthly) aggregated at the PRIO-GRID level

A dataset containing 5 multiple imputations of the UCDP Monthly aggregates at the PRIO-Grid level for those events that do not resolve to a precise PRIO-Grid. The methodology for imputations is described in Mihai Croicu and Håvard Hegre (2018) 'A Fast Spatial Multiple Imputation Procedure for Imprecise Armed Conflict Events (pdf)', Paper presented at the 59th Annual Convention International Studies Association, San Francisco, California.

A description of the method and data is available in the paper.
The ViEWS Outcomes Baseline Monthly codebook (pdf)
Imputed ViEWS Outcomes (Monthly) at the PRIO-Grid level (zip)

Please cite: Mihai Croicu and Håvard Hegre (2018) 'A Fast Spatial Multiple Imputation Procedure for Imprecise Armed Conflict Events', Paper presented at the 59th Annual Convention International Studies Association, San Francisco, California

Monthly PRIO-GRID forecasts, Mar 2019-Jan 2021

A dataset containing monthly forcasted predicted probabilities for 36 months (March 2019 - January 2021) at the PRIO-Grid level. Results from both individual component models and ensemble averages (average_select_sb, average_select_ns, average_select_os) are included in the file. See Hegre et. al. (2018) (pdf) for details.

Please cite: Hegre, Håvard, Marie Allansson, Matthias Basedau, Michael Colaresi, Mihai Croicu, Hanne Fjelde, Frederick Hoyles, Lisa Hultman, Stina Högbladh, Naima Mouhleb, Sayeed Auwn Muhammad, Desiree Nilsson, Håvard Mokleiv Nygård, Gudlaug Olafsdottir, Kristina Petrova, David Randahl, Espen Geelmuyden Rød, Nina von Uexkull, Jonas Vestby (2019) ‘ViEWS: A political violence early-warning system’, Journal of Peace Research, 56(2), pp. 155–174. doi: 10.1177/0022343319823860.

Monthly country forecasts, Mar 2019-Jan 2021

A dataset containing monthly forcasted predicted probabilities for 36 months (March 2019 - January 2021) at the country level. Results from both individual component models and ensemble averages (average_sb, average_ns, average_os) are included in the file. See Hegre et. al. (2018) (pdf) for details.