San Francisco Chapter of the American Statistical Association




Oct 17th from 4:30-6pm


Steven L Scott from Google


Predicting the Present with Bayesian Structural Time Series


Abstract: This presentation describes a system for short term forecasting based on an ensemble prediction that averages over different combinations of predictors. The system combines a structural time series model for the target series with regression component capturing the contributions of contemporaneous search query data. A spike-and-slab prior on the regression coefficients induces scarcity, dramatically reducing the size of the regression problem. Our system averages over potential contributions from a very large set of models and gives easily digested reports of which coefficients are likely to be important. We illustrate with applications to initial claims for unemployment benefits and to retail sales. Although our exposition focuses on using search engine data to forecast economic time series, the underlying statistical methods can be applied to more general short term forecasting with large numbers of contemporaneous predictors.


Location:  Benghazi Tech Talk, located in building 43 on Google Campus Address is 1600 Amphitheatre Parkway, Mountain View, CA 94043


Registration:    Attendees should email their name (consistent with the name on their ID that will need to be shown at the lobby), affiliation as well as contact email to the following email address: by Oct 13th.


Return to Bay Area ASA Homepage