Conflict Forecasting
Also known as: Political Violence Early Warning, Armed Conflict Forecasting, Conflict Early-Warning Systems, ViEWS-Style Forecasting
Conflict forecasting is the enterprise of producing calibrated, regularly updated probabilistic predictions of where and when armed conflict will occur, to support early warning and prevention. Exemplified by operational systems such as ViEWS (Hegre et al. 2019), it combines historical conflict data and predictors at fine spatial and temporal resolution, fits and ensembles multiple models, and forecasts violence months ahead — then rigorously evaluates those forecasts against what actually happens. It differs from explanatory conflict analysis by being transparent, prospective, and judged on out-of-sample accuracy rather than on coefficients.
Key highlights
- Produces actionable, regularly updated, probabilistic early warning of violence.
- Fine spatial and temporal resolution supports targeted prevention and response.
- Ensembling multiple models improves robustness and calibration over any single model.
- Transparent design and prospective out-of-sample evaluation build credibility and accountability.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use conflict forecasting when the objective is prospective early warning or risk assessment — telling decision makers where violence is most likely in the coming months — rather than explaining its causes. It suits humanitarian, diplomatic, and security planning. It is less appropriate when you need causal estimates or hypothesis tests, when data are too sparse or lagged to support honest forecasts, or when users would misread probabilistic risk as certainty. It builds on machine-learning prediction methods and on conflict event data such as UCDP and ACLED.
Strengths & limitations
- Produces actionable, regularly updated, probabilistic early warning of violence.
- Fine spatial and temporal resolution supports targeted prevention and response.
- Ensembling multiple models improves robustness and calibration over any single model.
- Transparent design and prospective out-of-sample evaluation build credibility and accountability.
- Forecasting rare, abrupt onsets and escalation remains hard; models predict continuation better than surprises.
- Dependence on reported data inherits reporting biases and lags.
- Calibrated probabilities can be misused or misunderstood by decision makers.
- Structural breaks and novel dynamics degrade models trained on the past.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How is conflict forecasting different from explanatory conflict analysis?
Explanatory analysis aims to identify and estimate the causes of conflict, judged by interpretable coefficients and identification. Forecasting aims to predict future conflict accurately, judged by prospective out-of-sample performance and calibration. A variable can be a strong cause yet a weak predictor, and vice versa, so the two enterprises use different models and evaluation standards even when they draw on the same data.
What makes a conflict forecast 'good'?
Prospective accuracy and calibration. It must predict genuinely future periods (not refit the past), discriminate well between high- and low-risk units under severe class imbalance (assessed with AUC and precision-recall), and be calibrated so that, say, cells assigned 20% risk experience conflict about 20% of the time. Transparency and regular updating also matter for operational credibility.
What is ViEWS and why is it influential?
ViEWS (the Violence Early-Warning System) is an operational, publicly available system, described by Hegre et al. (2019), that issues regularly updated probabilistic forecasts of organized violence at the country and PRIO-grid level for several months ahead. It is influential for combining transparency, fine resolution, ensemble modeling, and rigorous prospective evaluation, and for running open prediction competitions that have advanced the whole field.
Sources
- 1.Hegre, H., Allansson, M., Basedau, M., Colaresi, M., Croicu, M., Fjelde, H., et al. (2019). ViEWS: A political violence early-warning system. Journal of Peace Research, 56(2), 155–174.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 22). Conflict Forecasting. ScholarGate. https://scholargate.app/international-relations/conflict-forecasting