36-350 Information

Course Objectives

By the end of the course, you will be able to:
  1. Visualize data at multiple levels of detail and dimensionality
  2. Systematically remove layers of complexity from a dataset
  3. Estimate predictive relationships between variables, which may be continuous or discrete, involving regression or classification
  4. Implement these procedures in software, using S-plus

Course Procedure

There is no required text for the course. Notes and background readings for each lecture will be posted on the Web.

Homework will be assigned every week, including a mix of problem solving and computing. Reading and homework should take about six hours per week, and I will try to stick to that. Assignments will be posted on the web.

We will use S-plus for computing. S-plus is the premier language for statistical computing and is also used for data mining (read about the "Analytic Server" on Insightful's web page). S-plus is widely available on campus computers (type Splus) and is compatible with a free package called R. The assignments will be structured to help you learn the language as you go along. Later assignments will involve increasingly complex use of S-plus.

Grading policy

Homework is worth full credit at the beginning of class on the due date, usually Friday.
It is worth half credit until the beginning of class on Monday.
It is worth zero credit after that.

The lowest homework grade will be dropped except if it is the last assignment of the semester which is mandatory. The remaining homework grades will be used to compute the homework average. The homework average will contribute 75% to the course grade.

There will be a comprehensive final exam at the end of the course, during final exam period. The final examination will contribute 25% to the course grade.

The written description and interpretation of data analyses is at least as important as generation of data summaries, statistics, tests, etc. Grading of both homework and exams will take into consideration written interpretations as well as numerical accuracy.

All work and computer code must be your own.
See the CMU Student Handbook on Cheating and Plagiarism.

Tom Minka
Last modified: Sun Jan 13 16:42:00 EST 2002