Cluster-based early warning indicators for political change in the contemporary levant

We we cluster analysis to develop a model of political change in the Levant as reflected in the World Event Interaction Survey coded event data genera ted from Reuters between 1979 and 1998. A new statistical algorithm that us es the correlation between dyadic behaviors at two times identities cluster s of political activity. The transition to a new cluster occurs when a poin t is closer in distance to subsequent points than to preceding ones. These clusters begin to "stretch" before breaking apart, which served as an early warning indicator. The clusters correspond well with phases of political b ehavior identified a priori. A Monte Carlo analysis shows that the clusteri ng and early waltzing measures are not random; they perform very differentl y in simulated data: sets with similar statistical characteristics. Our stu dy demonstrates that the statistical analysis of newswire reports can yield systematic early warning indicators, and it provides empirical supper? for the theoretical concept of distinct behavioral phases in political activit y.