Predicting (and Understanding) the 2012 Election


July 26, 2012


David Rothschild


Microsoft Research


For more than 75 years, elections forecasting has been static—ask a random sample of a representative group of voters who they would vote for if the election were held today, and then report the poll result. In this talk, David Rothschild first demonstrates that the same samples could be addressed with other questions to produce a more accurate standard forecast (such as binary winner or expected vote share). Second, he challenges the standard forecast, stating that what most stakeholders really want and need are more innovative forecasts, like probability of victory or even probability distributions. Third, David shows how both standard and innovative forecasts can be made more efficient with new methods that utilize more cost-effective, non-representative samples and, in time, passively generated social media data. Fourth, he shows how Microsoft is going to be a leader in this new innovation. Finally, David tells you who is going to win the election!


David Rothschild

David Rothschild is an economist at Microsoft Research in New York City. He has a PhD in applied economics from the Wharton School of Business at the University of Pennsylvania. His dissertation is in creating aggregated forecasts from individual-level information. He has written extensively, in both the academic and popular press, on polling, prediction markets, and predictions of upcoming events. Most of his popular work has focused on predicting elections and an economist’s take on public policy. He has academic papers that cover political economy, behavior economics/public opinion, public economics/public policy, industrial organization, and experimental economics. Since joining Microsoft in May, he has been busy building prediction and sentiment models and organizing novel/experimental polling and prediction games.