MIXED LOGIT ESTIMATION OF RADICAL ISLAMIC TERRORISM IN EUROPE AND NORTH AMERICA: A COMPARATIVE STUDY

This paper estimates what characteristics (e.g. location, number of casualties, and type of attack) are associated with an Islamic terrorist attack. This is established by identifying the significant determinants of the probability that an attack had been carried out by Islamic terrorists. For Europe, the United States, and Canada, the analysis employs ITERATE data from October 1979 to December 2002 to ascertain the significant characteristics of Islamic terrorist attacks. A random- parameter logit model is used to analyze the probability of such attacks, taking into account the heterogeneity of the sample data. This model outperforms the standard logit model. Some policy implications are presented.