I study the causes and consequences of international conflict, including international crises, disputes, and wars. I employ statistical and simulation techniques to analyze the relationship between historical conditions (economic, political, and international) and the outbreak and continuation of international conflict. I often use rational, although not necessarily fully-developed game-theoretic formal, models to generate hypotheses for testing. I also use agent-based simulation to study emergent behaviors in a system of conflictual actors.

Some projects that I have completed include a comparative test of theories of international dispute outbreak and escalation (with Allan Stam), published in The Behavioral Origins of War (2004, University of Michigan Press). I have published analyses of war duration (with Allan Stam), alliance duration, and rivalry termination, and written about research design issues in the quantitative analysis of international relations. Work on The Behavioral Origins of War project included development of a software program (EUGene) which can 1) generate expected utility for all dyad-years since 1816, and 2) create a data set based on user-specifications for the unit of analysis (country-year, dyad-year), cases (e.g. all dyads, all major power dyads, all politically-relevant dyads), and variables (e.g. capabilities, tau-b scores, distance, expected utility, dispute occurence). This program is available at http://eugenesoftware.org. I have explored the relationship between economic conditions, democracy, power, and interstate rivalry termination.

At the outset of the 2003 Iraq-US war, Allan Stam and I published a prediction of the duration of the war. Those results were presented in an online research report, and have now appeared in the journal Foreign Policy Analysis (full replication data and forecasting spreadsheets are available under the "datasets" page).

Other recent projects include a study of the democratic peace which better maps out the empirical relationship between democratic similarity, autocratic similarity, and conflict. I continue to model the relationship between military strategy, civilian casualties, and the spread of insurgency using agent-based models developed using the Repast simulation libraries. I am expanding a baseline model of insurgent-government interactions (originally published 2008) to add the possibilities of agent learning, the presence of multiple insurgent groups who may compete with one another as well as the government, and peaceful recruitment as a viable option for both government and insurgent agents. With co-authors,I have examined empirical indicators that may allow us to assess the accuracy and effectiveness of government agents to move towards broad empirical testing of the model. In contexts beyond insurgency, I have also explored combining an agent-based model of international action with rational choice models as a basis for decision, and also linking empirical data on the conduct of war and diplomacy to better integrate agent-based models with empirical reality. The goal is to make such models more accessible (a la EUGene), and also more useful for theory/hypothesis development with clear links to testing.

I teach courses in the area of international relations. Courses I have taught include the introductory undergraduate "Introduction to International Politics," and the graduate international relations proseminar, "International Relations Theory and Methodology." I also teach the undergraduate courses "International Relations Theory" and "War in World Politics," and graduate level research seminars "International Conflict" and "Conflict Resolution and Bargaining."