stats306B - applied multivariate statistics


Instructor:
 Prof. J. Taylor
Sequoia Hall #137
Email
723-9230
Schedule:
 TTh 12:50-2:05
Location:
 Sequoia Hall #200
Office Hours:
 TTh 2:30-3:30, or by appointment
References:
TAs & Office Hours:
  • Genevera Allen, TTh 11:50-12:50, Sequoia Hall #206
  • Camilo Rivera, WF 1:00-2:00, Sequoia Hall #227.
Computing:
  • I will use python for most of the examples. See here for more resources.
  • All necessary libraries will be installed on miller, rupert.
  • Documentation for the python examples is here.
  • Windows users can install enthon which has (almost) everything we need. Here is a local copy.
Scheduling:
 I have to be away on April 8, May 1. We will have to reschedule these courses.
General
Outline:
 I will follow Multivariate analysis for much of the course (adding material as needed), focusing on (parts of) Chapters 3, 8-14.
  • Some multivariate normal theory.
  • Principal component analysis.
  • Factor analysis.
  • Independent components analysis.
  • Discriminant analysis.
  • Cluster analysis.
  • Multidimensional scaling.
  • Manifold learning.
Prerequisites: A regression course, such as STATS 305.
Evaluation:
  • 3 assignments: 75%
  • final exam: 25%
Notes
Assignments:
 You may discuss homework problems with other students, but you should prepare the written assignments by yourself. Please show your work. We will count days late on each problem set. Each day late is penalized by 10% of the homework value. Homework more than 3 days late will ordinarily get 0. If you're travelling, you can email a pdf file. For sickness, interviews and other events, up to 3 late days total are forgiven at the end of the quarter. (Work late enough to get zero does not get redeemed though.)