A0154
Title: Statistical contributions in conflict research
Authors: Goeran Kauermann - LMU Munich (Germany) [presenting]
Abstract: The field of statistical conflict research is showcased. Armed conflicts are considered in regions of Africa with at least one fatality. The reliable data provided by the Peace Research Institute Oslo (PRIO) is used as a data source. Firstly, the way in which statistical models can be used to predict future conflicts is demonstrated. A hurdle model is applied, which allows the proper incorporation of quantities relevant from the political science perspective. Secondly, Syria is considered and explored in terms of how remote sensing data from satellites can improve prediction accuracy. In particular, machine learning tools are included, and those are compared with statistical approaches. Finally, the spread of conflicts is looked at in time and space, which is often omitted in purely machine learning approaches.
