CSI 991

Seminar in Computational Statistics:

Statistical Learning

Spring, 2002

Fridays 3:30pm -- 5:00pm, Science & Technology, Room 206

We will work through some of the material in The Elements of Statistical Learning, Data Mining, Inference and Prediction by Hastie, Tibshirani, and Friedman (HTF).
Website for text.

We will also do some exploratory analyses in S-Plus, CART and MARS.
Additional errata.

Schedule

January 18
Organizational meeting. Preliminary discussions of HTF and other texts and articles.
January 25
Discuss Chapter 5. Comments and example S-Plus commands for spline smoothing.
February 1
Discuss Breiman's interview in Statistical Science.
February 8
Discuss Breiman's ``two cultures'' article in Statistical Science.
February 15
Discuss Chapter 6. (Each participant was assigned a section to lead.)
Jim Shine will look into getting Breiman's code for boosting incorporated into R.
February 22
Discuss Chapter 6.
March 1
Discuss Chapter 7.
March 8
Discuss Chapter 7.
March 15
No meeting
March 22
Discuss Chapter 7.
March 29
Discuss Chapter 8.
April 5
Discuss Chapter 8.
April 12
Finish Chapter 8.
April 19
Discuss Section 9.1 and 9.2
April 26
Discuss Section 9.3 and 9.4
Experiment with MARS using canned data, and then generate some data for further experimentation.
Jim Shine and Jon Schuler will look into getting the PRIM code incorporated into S-Plus.
May 3
Continue experimenting with MARS.

Here's a simple exercise:
Find the groups in this dataset.

Links to some classification software.