Skip to main content Accessibility help

Self-exciting point process models for political conflict forecasting

  • N. JOHNSON (a1), A. HITCHMAN (a1), D. PHAN (a1) and L. SMITH (a1)


In 2008, the Defense Advanced Research Project Agency commissioned a database known as the Integrated Crisis Early Warning System to serve as the foundation for models capable of detecting and predicting increases in political conflict worldwide. Such models, by signalling expected increases in political conflict, would help inform and prepare policymakers to react accordingly to conflict proliferation both domestically and internationally. Using data from the Integrated Crisis Early Warning System, we construct and test a self-exciting point process, or Hawkes process, model to describe and predict amounts of domestic, political conflict; we focus on Colombia and Venezuela as examples for this model. By comparing the accuracy of fitted models to the observed data, we find that we are able to closely describe occurrences of conflict in each country. Thus, using this model can allow policymakers to anticipate relative increases in the amount of domestic political conflict following major events.



Hide All
[1] Andriole, S. & Young, R. (1977) Toward the development of an integrated crisis early warning system. In: International Studies Quarterly, pp. 107–50.
[2] Atran, S. & Ginges, J. (2012) Religious and sacred imperatives in human conflict. Science 336 (6083), 855–57.
[3] Barathy, G. & Silverman, B. (2012) Holistically evaluation agent-based social systems models: A case study. Simulation 89 (1), 102135.
[4] Bond, D., Jenkins, J., Taylor, C. & Schock, K. (1997) Mapping mass political conflict and civil society issues and prospects for the automated development of event data. J. Conflict Resolution 41 (4), 553579.
[5] Boshcee, E., Natarajan, P. & Weischedel, R. (2013) Automatic extraction of events from open source text for predictive forecasting. In: Handbook of Computational Approaches to Counterterrorism, pp. 51–67.
[6] Brandt, P., Freeman, J. & Schrodt, P. (2011) Real time, time series forecasting of inter-and-intra-state political conflict. In: Conflict Management and Peace Science, pp. 41–64.
[7] Cox, D. & Isham, V. (1980) Point Processes, CRC Press, London.
[8] Egesdal, M., Fathauer, C., Jouie, K., Neuman, J., Mohler, G. & Lewis, E. (2010) Statistical and stochastic modeling of gang rivalries in Los Angeles. SIAM Undergraduate Research Online 72–94.
[9] Freeman, J. & Job, B. (1979) Scientific forecasts in international relations: Problems of definition and epistemology. International Studies Quarterly 23 (1), 113–43.
[10] Germer, D., Schrodt, P., Yilmaz, O. & Abu-Jabr, R. (2002) Conflict and mediation event observation (CAMEO): A new event data framework for the analysis of foreign policy interactions. International Studies Association.
[11] Goldstein, J. (1992) A conflict-cooperation scale for weis events data. J. Conflict Resolution 36 (2), 369385.
[12] Helmstetter, A. & Sornetter, D. (2003) Importance of direct and indirect triggered seismicity in the etas model of seismicity. Geophys. Res. Lett. 30 (11), 15761579.
[13] Lewis, E., Mohler, G., Brantingham, P. & Bertozzi, A. (2012) Self-exciting point process models of civilian deaths in Iraq. Secur. J. 25 (3), 244264.
[14] Mahoney, S., Comstock, E. & Darcy, S. (2011) Aggregating forecasts using a learned bayesian network. In: Twenty-Fourth International FLAIRS Conference.
[15] Mohler, G., Short, M., Brantingham, P., Schoenberg, F. & Tita, G. (2012) Self-exciting point process modeling of crime. J. Am. Stat. Assoc. 106 (493), 100108.
[16] Ozaki, T. (1979) Maximum likelihood estimation of Hawkes' self-exciting point processes. Ann. Inst. Stat. Math. 31 (1) 145155.
[17] O'Brien, S. (2013) A multi-method approach for near real time conflict and crisis early warning. In: Handbook of Computational Approaches to Counterterrorism pp. 401–418.
[18] Perkel, D., Gerstein, G. & Moore, G. (1967) Neuronal spike trains and stochastic point processes: the single spike train. Biophys. J., 7 (4), 391418.
[19] Racette, M., Smith, C., Cunningham, M., Heekin, T., Lemley, J. & Mathieu, R. (2014) Improving situational awareness for humanitarian logistics through predictive modeling. In: Systems and Information Engineering Design Symposium (SIEDS), pp. 334–339.
[20] Roberts, K. (2003) Social correlates of party system demise and populist resurgence in Venezuela. Latin America Politics and Society 45 (3), 3557.
[21] Schrodt, P., Yilmaz, O., Gerner, D. & Hermreck, D. (2008) The CAMEO (conflict and mediation event observations) actor coding framework. 2008 Annual Meeting of the International Studies Association.
[22] Sheather, S. (2004) Density estimation. Statistical Science 19 (4), 588597.
[23] Silverman, B. (1986) Density Estimation for Statistics and Data Analysis, CRC Press, London.
[24] Singer, J. & Wallace, M. (1979) To augur well: Early warning indicators in world politics. Sage Publications, First edition.
[25] Snyder, D. & Miller, M. (1991) Self-exciting point processes. Second edition. Random Point Processes in Time and Space, 287–340.
[26] Tench, S., Fry, H. & Gill, P. (2016) Spatio-temporal patterns of IED usage by the provisional Irish Republican Army. Eur. J. Appl. Math. 27 (3), 377402.
[27] Terrell, G. & Scott, D. (1992) Variable kernel density estimation. Ann. Stat. 1236–1265.
[28] Waisberg, T. 2008 Colombia's use of force in Ecuador against terrorist organization: International law and the use of force against non-state actors. Am. Soc. Int. Law. 12 (17).
[29] Ward, M., Beger, A., Cutler, J., Dickenson, M., Dorff, C. & Radford, B. (2013) Comparing GDELT and ICEWS event data. Analysis, 21 (1) 267–97.
[30] Ward, M., Metternich, N., Carrington, C., Dorff, C., Gallop, M., Hollenbach, F., Schultz, A. & Weschle, S. (2012) Geographical models of crises: Evidence from ICEWS. Adv. Design Cross-Cultural Act. 429.


Self-exciting point process models for political conflict forecasting

  • N. JOHNSON (a1), A. HITCHMAN (a1), D. PHAN (a1) and L. SMITH (a1)


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed