O'Brien: Anticipating the Good, the Bad and the Ugly

O'Brien, S., 2002. Anticipating the Good, the Bad, and the Ugly: An Early Warning Approach to Conflict and Instability Analysis. Journal of Conflict Resolution, 46(6), 791.

This study uses a statistical technique called fuzzy analysis of statistical evidence (FASE) to determine whether or not there are country-level, “macrostructural factors that can contribute to different kinds and levels of intensity of conflict and country instabilities” (791). O’Brien concludes that his method can forecast both the occurrence and intensity of conflicts for the following 5 years with 80% accuracy.

The paper begins by exploring the need for policy relevant scholarship, especially that exploring early-warning systems. The paper highlights a variety of studies that attempt to provide such scholarship. It then explores the State Failure Task Force (SFTF) and their four types of state failure: genocides/politicides, ethnic wars, revolutionary wars, and adverse/disruptive regime transitions (792).

The SFTF used a variety of methodological tools, including logistic regressions, neural networks and genetic algorithms to look at patterns in structural data. They found that their best model included only three variables: level of democracy, trade openness and infant mortality rate (793). They were about 66% effective with these variables historically. King and Zeng outperform the SFTF model by including legislative effectiveness and fraction of population in the military.

This study attempts to expand upon these earlier approaches.

He uses KOSIMO data

TRUNCATED HERE