This page describes in general terms the variables and techniques we have used to produce our forecasts. Users interested in full details should consult our academic papers listed in publication section of this website.
We build on previous analysis of genocide/politicide, mass killing, civil war, and political instability to inform our forecasting model. We develop a two-stage approach which considers genocide and politicide as emerging from already unstable political situations. We use variables found to be significantly associated with instability and/or genocide/politicide in previous studies, but we give preference to the forecasting power of each factor, rather than statistical significance. Informally we define politicide as mass killing intended to eliminate a group in whole or in part for groups which are identified politically (e.g., by class or by ideology, such as bourgeois elements in Cambodia or suspected Communists in Indonesia). Thus genocide refers to mass killing of the same sort in which the target group is defined ethnically (e.g., Jews, Tutsis, or Armenians). Our formal definitions and data are taken from the work of the Political Instability Task Force. Our models include a range of variables measuring historical, political-economic, and highly time-sensitive factors.
A Stylized Picture of Our Model
For our post-Cold War forecasts we have also added the following variables:
- Peacekeeping forces in-country
- Guerrilla tactics used in-country
- Internationalized internal conflict
- Country involved in an interstate conflict
- Linear and non-linear terms to capture time-trends not otherwise measured in our data