Carnegie Mellon University
Skip navigation and jump directly to page content

Dec. 3: CMU Statistics Team Receives Grant To Analyze Effectiveness of Economic Forecasting Models

Contact: Shilo Raube / 412-268-6094 /

CMU Statistics Team Receives Grant To Analyze
Effectiveness of Economic Forecasting Models

Current Predictors Failed To Foresee Financial Collapse

PITTSBURGH—Three Carnegie Mellon University statisticians are working to find out why the dominant economic models used by economists for forecasting and setting public policy failed to predict the recent financial collapse or even to understand the recession's course after the fact.

Cosma Shalizi, Mark Schervish and Daniel McDonald, have received a three-year $144,814 grant from the Institute for New Economic Thinking (INET) to use proven statistical learning techniques to evaluate the effectiveness of economic models. Their work will allow economists to confidently analyze models and help them decide which models to use to make predictions.

"Economists have to use models for forecasting and policy-making, but currently they do not have anything to tell them that the models are any good," said Shalizi, assistant professor of statistics. "INET is giving us the opportunity to bring model evaluation up to the same rational level everyone uses in data mining and statistical learning. In the end, we'll show economists how they can reliably select the best models and control prediction error."

Shalizi, Schervish, head of the Department of Statistics, and McDonald, a statistics Ph.D. candidate, have deep experience in the statistical analysis of complex systems, including time series prediction, self-organization and network analysis. The team plans to use INET's funding to resolve macroeconomic disputes and determine the reliability of models that emerge for macroeconomic time series. They will do this by adapting machine-learning techniques that control overfitting to economic models.

"Over the last three decades, statisticians and computer scientists have developed sophisticated methods of model selection and forecast evaluation, under the rubric of statistical learning theory," said Robert Johnson, executive director of INET. "Applying the methods that have revolutionized the modern industry of data mining is an approach that we believe holds great promise in improving the quality of economic forecasts and predictions."

INET was launched with a $50 million pledge from George Soros to promote changes in economic theory and practice through research grants, task force groups and academic partnerships. The CMU grant was established through the institute's Inaugural Grant Program.

For more information on the Carnegie Mellon Statistics Department, visit