Dr. Joseph J. Retzer, University of Wisconsin-Milwaukee
MARS (Multivariate Adaptive Regression Splines) is a modern, computer intensive data-mining technique developed by Jerome H. Friedman of Stanford University. MARS is useful for handling a variety of challenges when modeling data among them being the identification of non-linear relationships between variables.
This talk will begin with a short description of a few common techniques for discovering relationships between variables along with their limitations. Next, the MARS approach will be introduced and illustrated. Results from an application of MARS to actual data will also be presented.
Bio: Dr. Retzer has over 20 years of experience in market research analytics. During this time he has developed innovative statistical techniques in areas including key driver measurement in the presence of collinearity, prediction in covariance structure models, modeling of behavioral loyalty using survival analytic techniques, genetic algorithm based segmentation and Bayesian inference. His research interests include applied statistical and econometric analysis of marketing models in both classical and Bayesian frameworks.
Meetup Sponsor: Cloudera
5:15pm - 6:00pm - Networking, Food and Beverages
6:00pm - 7:30pm - Presentation
7:30pm - 8:00pm - Networking
Attendees must register at the first floor lobby. The meetup will be held in the lower level conference room.
Parking: Use this link to the Marquette Campus Map for near-by parking. Note our venue, Raynor Memorial Library, is building 20 on the map. The Well's street parking ramp (67 on map) is of ample size and about a two block walk from Raynor Memorial Library.