Self-exciting Hawkes' process for terrorism activities

Details

Supervisors

Hainaut, Donatien

Faculty

Faculté des sciences

Degree label

Master [120] en sciences actuarielles, à finalité spécialisée

Abstract

Terrorism activity is an important topic for (re-)insurance company as it is a risk that couldimpact life, property and infrastructure. It’s a risk which is characterized by the fact that it is arare event with a huge impact when it occur. In Belgium, the Terrorism Reinsurance InsurancePool (TRIP) is a non-profit organization created by Assuralia and other insurance company forthe purpose of compliance with the Law of 1 April 2007 on insurance against damage caused byterrorism. Here is the definition of a terrorism act given by the TRIP : "Terrorism is an act orthreatened act organized in secret for ideological, political, ethnic or religious ends, performedindividually or in groups and intended as an attempt at the lives of individuals or to eitherpartially or completely destroy the economic value of tangible or intangible property whether toimpact on the public, create a climate of insecurity or put pressure on the authorities in a bidto impede the running and normal operation of a service or business. " This definition will beimportant for the data that will use for the model The goal of this thesis will be to use some mathematical, statistical and actuarial tools in order to increase the understanding of terrorism activity within a country. A very important hypothesis in terrorism activity is that the attacks are clustered which means that past events have an influence on current events. The consequence of this hypothesis is that standard models which assume independence are unsuitable. This is the main reason for which Self-exciting Hawkes' process will be used in order to confirm or not this hypothesis. Clustering effects and contagion behaviour has been demonstrated in international terrorism [Midlarsky, Crenshaw andYoshida 1980].