A Brief History of Classification and Regression Trees
Speaker: Wei-Yin Loh, University of Wisconsin - Madison
Chair: Mike Fleming
Date and Time: June 4, 2015, 12:30pm to 1:30 p.m.
Location: Bureau of Labor Statistics Conference Center
Location: Bureau of Labor Statistics Conference Center
Sponsor: Methodology Section
Slides: loh_slides.pdf
Abstract:
It has been 52 years since the publication of AID (Morgan and Sonquist, 1963), the first regression tree algorithm. Owing to its novelty at the time and to the many questions it left unanswered, the paper was met with mostly criticism and cynicism. As a result, little progress was made until the appearance two decades later of CHAID (Kass, 1980), CART (Breiman et al., 1984), ID3 (Quinlan, 1986), and FACT (Loh and Vanichsetakul, 1988). These second-generation methods brought new ideas that resolved many of the issues of AID, but they also revealed some fundamental computational and statistical problems of their own. Research accelerated over the last two decades, being spurred almost simultaneously by the rapid increase in computing power, the rapid growth in the size and complexity of data sets, and the desire for more flexible statistical techniques. This talk will give a brief critical review of the history of the subject and the current state-of-the-art algorithms.
POC e-mail: loh@stat.wisc.edu
WebEx event address for attendees:
https://dol.webex.com/dol/j.php?MTID=m367db7dbfefd17af098f42d2aa056d9f
For audio: Call-in toll-free number (Verizon): 1-866-747-9048 (US)
Call-in number (Verizon): 1-517-233-2139
(US) Attendee access code: 938 454 2
