Predicting Local Violence

December 19, 2014

Investigators: CDEP Affiliate Chris Blattman and Robert Blair and Alexandra Hartman from Yale University.

This research tests the feasibility of local-level violence forecasting. It applies standard prediction models to new data from 242 Liberian communities to investigate whether it is possible to predict outbreaks of local violence with sensitivity and accuracy, even with limited data. First, the models are trained to predict communal, extrajudicial, and criminal violence in 2010 using 2008 risk factors. Then, forecasts are made of violence in 2012, before collecting data. The model predicts up to 88% of actual 2012 violence. This comes at the cost of many false positives, for overall accuracy of 33 to 50%. Policy-wise, states and peacekeepers could use such predictions to prevent and respond to violence. The models also generate new stylized facts for theory to explain. In this case, ethnic cleavages and power-sharing predict violence, while economic variables typically do not. The researchers illustrate how forecasting can be widely more applied to micro-level conflict data.

Published as "Predicting Local Violence" (April 15, 2015).