Forests of Uncertaint(r)ees: Using tree-based ensembles to estimate probability distributions of future conflict

TL;DR

This paper develops a machine learning approach using tree-based ensembles to predict and quantify the uncertainty of future conflict fatalities at a granular spatial level, improving forecast reliability.

Contribution

It introduces a novel AutoML framework combining classifiers and distributional regressors to generate full predictive distributions for conflict outcomes, incorporating regional models to reduce uncertainty.

Findings

Models outperform benchmarks in one-year-ahead conflict predictions.

Integration of regional models does not decrease predictive performance.

Models accurately reflect meaningful performance improvements in conflict-affected regions.

Abstract

Predictions of fatalities from violent conflict on the PRIO-GRID-month (pgm) level are characterized by high levels of uncertainty, limiting their usefulness in practical applications. We discuss the two main sources of uncertainty for this prediction task, the nature of violent conflict and data limitations, embedding conflict prediction in the wider literature on uncertainty quantification in machine learning. Based on this, we develop a strategy to quantify uncertainty in conflict forecasting, shifting from traditional point predictions to full predictive distributions. Our approach combines multiple tree-based classifiers and distributional regressors in a custom AutoML setup, estimating distributions for each pgm individually. We also test the integration of regional models in spatial ensembles as a potential avenue to reduce uncertainty by lowering data requirements and accounting for systematic differences between conflict contexts. The models are able to consistently outperform a suite of benchmarks derived from conflict history in predictions up to one year in advance. Marginal differences in model-wide metrics emphasize the need to understand their behavior for a given prediction problem, in this case characterized by extremely high zero-inflatedness. Adressing this, we compliment our evaluation with a simulation experiment, which demonstrates that our models reflect meaningful performance improvements, which can be traced back to conflict-affected regions. Lastly, we show that the integration of regional models does not decrease performance, opening avenues to integrate additional data sources in the future.