A shape-based approach to conflict forecasting

Abstract

Do conflict processes exhibit repeating patterns over time? And if so, can we exploit the recurring shapes and structures of the time series to forecast the evolution of conflict? Theory has long focused on the sequence of events that precedes conflicts (e.g., escalation or brinkmanship). Yet, current empirical research is unable to represent these complex interactions unfolding over time because it attempts to match cases on the raw value of covariates, and not on their structure or shape. As a result, it cannot easily represent real-world relations which may, for example, follow a long alternation of escalation and détente, in various orders and at various speeds. Here, I aim to address these issues using recent machine-learning methods derived from pattern recognition in time series to study the dynamics of casualties in civil war processes. I find that the methods perform well on out-of-sample forecasts of the count of the number of fatalities per month from state-based conflict. In particular, our results yield Mean Squared Errors that are lower than the competition benchmark. We discuss the implication for conflict research and the importance of comparing entire sequences rather than isolated observations in time.