Prediction Engines


July 15, 2013


Robin Hanson and Sanmay Das


Virginia Polytechnic Institute & State University, George Mason University


From crowdsourcing to data-driven models, technology offers new ways to collect and aggregate information on an unprecedented scale, allowing researchers to make reliable predictions about elections, policy, corporate decisions, economics, finance, sports, and entertainment.

This session of the 2013 Microsoft Research Faculty Summit reports on progress in the science and engineering of polls, prediction markets, and forecast models. Examples include a fundamental model that correctly predicted 49 of 50 states nine months before the US Presidential election, and combinatorial prediction markets capable of estimating billions of complex predictions like correlations between states.


Robin Hanson and Sanmay Das

Sanmay Das is an associate professor of Computer Science at Virginia Tech. He was previously an assistant professor in the Computer Science Department at Rensselaer Polytechnic Institute, where he also held a courtesy appointment at the Lally School of Management. His research interests are in machine learning and computational social science. He received an NSF CAREER award in 2010. He has served as program chair of AMMA and workshops chair of ACM EC, in addition to serving on the program committees of many conferences in artificial intelligence and machine learning. Sanmay received his A.B. from Harvard and his Ph.D. from the Massachusetts Institute of Technology, both in Computer Science.

Robin Hanson is an associate professor of economics at George Mason University, a research associate at Oxford’s Future of Humanity Institute, and chief scientist at Consensus Point. After receiving his 1997 social science Ph.D. from Caltech, he was a Robert Wood Johnson Foundation health policy scholar at the University of California at Berkeley. In 1984, Hanson received a master’s degree in physics and a master’s degree in the philosophy of science from the University of Chicago, and afterward spent nine years researching artificial intelligence, Bayesian statistics, and hypertext publishing at Lockheed, NASA, and independently. Since 1988, he has pioneered prediction markets, in other words, speculative markets subsidized to gain better estimates on important topics. Hanson was a principal architect of the first internal corporate markets and the first web markets. He has developed new technologies for conditional, combinatorial, and intermediated trading, and has studied insider trading, manipulation, and other foul play.