Self-exciting hurdle models for terrorist activity

TL;DR

This paper introduces a self-exciting hurdle model to predict terrorist attacks, capturing the influence of past attacks and the occurrence of multiple attacks on the same day, with application to Indonesia's data from 1994-2007.

Contribution

It develops a novel self-exciting hurdle model combining shot noise processes and power law distributions for terrorist activity prediction.

Findings

Model accurately predicts attack probabilities and counts.

Power law distribution best fits attack numbers.

Model parameters provide insights into attack dynamics.

Abstract

A predictive model of terrorist activity is developed by examining the daily number of terrorist attacks in Indonesia from 1994 through 2007. The dynamic model employs a shot noise process to explain the self-exciting nature of the terrorist activities. This estimates the probability of future attacks as a function of the times since the past attacks. In addition, the excess of nonattack days coupled with the presence of multiple coordinated attacks on the same day compelled the use of hurdle models to jointly model the probability of an attack day and corresponding number of attacks. A power law distribution with a shot noise driven parameter best modeled the number of attacks on an attack day. Interpretation of the model parameters is discussed and predictive performance of the models is evaluated.