andreas-graefe.org

  • Increase font size
  • Default font size
  • Decrease font size
Andreas Graefe

About me

I am a researcher and doctoral student at the Institute for Technology Assessment and Systems Analysis at the Research Center (Forschungszentrum) Karlsruhe, Germany. Since January 2008, I'm visiting with Professor Scott Armstrong at the University of Pennsylvania's Wharton Business School.

I received a diploma (German equivalent to a master's degree) in Economics and a diploma in Information Science from the University of Regensburg (Germany) in December 2004 and March 2005, respectively. Between 2001 and 2002 I also studied for two semesters at the University of Zurich (Switzerland).

In my dissertation, which is supervised by Prof. Christof Weinhardt at the chair for Information and Market Engineering at the University of Karlsruhe, I'm comparing prediction markets to traditional judgmental forecasting methods, in particular prediction markets and the Delphi method. I'm also running the Special Interest Group on Prediction Markets at forecastingprinciples.com.

 

PollyIssues

Together with Scott Armstrong, I'm working on a model to predict the outcome of the U.S. presidential elections based on how voters perceive the candidates to deal with the issues facing the country. For 2008, our model predicts Barack Obama to win, with a vote-share of 52.5%. The work has already achieved some media attention, particularly in Germany.

Further information can be found here.

 

Special Interest Group on Prediction Markets

IIF logoThe marketsforforecasting.com Special Interest Group on prediction markets at forecastingprinciples.com has been established as a platform for researchers and practitioners who are interested in the field of prediction markets.
 

PollyVote – Forecasting the US Presidential Election

Polly is interested in forecasting the percent of the two-party vote going to the incumbents, the Republican candidate (whoever that might turn out to be). Her procedure is to average across four forecasting methods (polls, the IEM prediction market, quantitative models, and her own Experts Panel).

Further information can be found on PollyVote.com.