skip to content

Faculty of Economics

Thursday, 25 March, 2021

Forecasting a phenomenon like armed conflict up to a year in advance is difficult, and history has shown conflict might suddenly appear out of a previously long lived and low level dispute.

Dr Christopher Rauh, a University Lecturer at the Faculty of Economics at the University of Cambridge, is part of a team that has investigated the signals that indicate a conflict might be about to turn ‘hot’, and result in civil commotion turning into either an all out war, or at least a dispute that would result in fatalities.

“A special feature of our model is our reliance on newspaper text. This allows us to capture changes in low level risk of conflict in countries which do not have a recent history of violence,” he says. “The risk estimates provided here are ambitious in that they put an emphasis on these cases.”


The models provide risk forecasts for two levels of violence:

  1. An outbreak of any violence: a country goes from no fatalities to positive fatalities.

  2. An outbreak of armed conflict: a country goes from having less than 50 fatalities in a month to more than 50 fatalities in a month.


“Our model has been tested extensively by us and in a forecast competition organized by the Violence Early-Warning System (ViEWS) in which our approach performed extremely well,” he adds.

More details of the forecasting methods are at: