Statistics 306A: (Winter 2001)

   Methods of applied statistics

TIME: MW 12:50 - 2:05pm

 Location: TCseq102

Instructor: Jerome H. Friedman

Logistics:

• Office hour: Thursday, 12:00pm, 134 Sequoia Hall.
• TA's: Sergiy Terentyev  (steren@stat.stanford.edu) & Xiaohu (Tom) Zang (zhangxh@stat.stanford.edu) .
• Text: Course reader: Modern Regression and Classification. Friedman, Hastie & Tibshirani.
• Homework:
  • traditional.
  • (journal) reading assignments (with/without questions).
  • computing assignments (apply / implement methods - discuss results).
• Midterm: maybe -doubt it.
• Final: likely - not sure what form.

Topics:

 I. Overview of statistical learning:

• Application areas
• Statistical model
• Types of variables
• Curse-of-dimensionality
• Regularization
• Dictionary methods
• Kernel methods
• Model selection

   II. Methodology:

• Smoothing and curve estimation
• Additive modeling
• Projection Pursuit regression
• Neural networks
• Radial basis functions
• CART/MART/MARS
• Support vector machines
• Selected topics: If time left: Bagging & boosting, association rules, clustering.