Abstract

Identifying armed group presence is often helpful for conflict studies to examine patterns of conflict. Armed group presence is often used as the main variable of interest in several studies, and in some cases, this variable is ignored. Many of these studies use expert data or proxy variables to analyze armed group presence. This paper proposes Hidden Markov Models (HMMs) as a method to identify armed group presence. HMMs permit identifying armed group presence at a sub-national level and in long panel data sets. A HMM is used in this paper to identify paramilitary and FARC presence in Colombia. The armed groups' presence predictions are used to analyze the effect of economic shocks on violence. It shows that economic shocks can have differential effects depending on the level of armed group presence in the territory.

Committee Chair

DR. NAN LIN

Committee Members

DR. TED ENAMORADO DR. JOSE FIGUEROA-LOPEZ

Degree

Master of Arts (AM/MA)

Author's Department

Statistics

Author's School

Graduate School of Arts and Sciences

Document Type

Thesis

Date of Award

Spring 5-2021

Language

English (en)

Download

Share

COinS