Andreas Beger

Data scientist

Taking over the world


Tags: #Crisp #Duke University #Icews #Prediction #Predictive Heuristics #Ward Lab

I work as a Postdoctoral Fellow in the Ward Lab here at Duke University. The Lab currently consists of Mike Ward, me, and a group of very smart graduate students. There are a lot of exciting projects within the lab, like ICEWS and other work for the US government, but also a broader set of projects by our lab members. One of the things we wanted to do this semester is to publicize this work a little bit more, and to this end we’re taking a new blog live today: Predictive Heuristics.

The idea is to have an outlet in which the lab as a whole and the people in it can present work in progress or other lab projects that might be of interest to what we call the predictive heuristics community: researchers and analysts who are using data and statistical methods to develop tools that aid real-world decisions-making and policy. To the extent that my own blog and career has an identity—military intelligence and statistical methods—that fits well with what I’m (sort of) trying to do. I think it will also be a nice way to keep people who are interested in ICEWS (which we are trying to make publicly available in some form) and other lab projects updated.

In any case, we’re going to have an interesting series of topics lined up. Although the lab as a whole is centered on predicting political conflict, we have people who are pursuing a pretty broad range of interests, from party and coalition politics through more traditional inter-state dynamics to imputation and other methods. Some of the posts we have coming up are on:

  • comparing the ICEWS and GDELT event data,
  • our collaboration with the Good Judgment project,
  • latent space modeling of Germany’s parties,
  • details of our global conflict forecasting R packages, ICEWS and CRISP.