Statistics 431: Statistical Inference
University of Pennsylvania
Fall 2006

Instructor
Jon McAuliffe
Office: 467 Huntsman Hall
Email: mcjon@wharton.upenn.edu
Office hours: Mondays and Wednesdays, 10:30 AM - 12:00 PM, and by appointment

Teaching assistant

Abhishek Gupta
Office: 414.1 Huntsman Hall
Email: abgupta@wharton.upenn.edu
Office hours: Thursdays, 9:00 AM - 12:00 PM, and by appointment

Syllabus:

• pdf

Announcements

• Abhishek will be available in his office from 2pm to 3pm tomorrow, Tuesday 12 December.
• Remarks on the practice final exam:
  • You do not need to know about Cook's distance or the Durbin-Watson test for the exam.
  • Problem 3B: The correct answer is (c), not (d) as reported in the solutions.
  • Problem 6G: The problem should have reported S(mu hat) for you to include in your PI calculation, and the solution should have used it. See Lecture 20, slide 5.
• Here is a practice final exam: pdf. In question (1), please ignore the "randomized block design," which we did not cover. Please skip questions (3)(G) and (3)(L), which relate to randomized block designs.
• Here is the solution to the practice final: pdf.
• Here is the facts and formulas sheet which will be made available to you at the final: pdf.
• The final examination will be Thursday 14 December, 9:00AM - 11:00AM, in G60 JMHH.
• *** LECTURE CANCELLED ***
  Lecture for Monday November 20th is cancelled. Please notify your fellow students. Since no lecture is scheduled for Wednesday November 22nd, our next meeting will be Monday November 27th.
  The problem set due on the 20th is now due on the 27th.
• When you take the midterm, this "facts and formulas" sheet will be provided to you: pdf. The midterm will cover all course materials up to (but not including) two-sample tests of population variances (Devore 9.5). The textbook discusses some ideas you will not be tested on:
  • The bootstrap
  • Tolerance intervals
  • One-sample inference about a population variance
  You should bring only a pen/pencil and calculator to the midterm. A copy of the facts and formulas sheet will be provided once you arrive.
• Here is a practice midterm exam: pdf
• The due date for problem set 2 has been moved from Monday 2 October to Wednesday 4 October. The problem set is still due at the beginning of lecture, 1.30pm. Problem set 3 will still be due Monday 9 October.
• Some people experienced problems with the lecture 5 pdf file. I have put a new version in place. Please email me if you have continuing problems.

Problem sets

• Problem set 01, due Mon 25 Sep, 1:30 PM: pdf
• Problem set 02, due Wed 04 Oct, 1:30 PM: pdf
• Problem set 03, due Mon 09 Oct, 1:30 PM: pdf
• Problem set 04, due Mon 16 Oct, 1:30 PM: pdf
• Problem set 05, due Mon 30 Oct, 1:30 PM: pdf     [ handout on one-way ANOVA in JMP: pdf ]
• Problem set 06, due Mon 06 Nov, 1:30 PM: pdf
• Problem set 07, due Mon 13 Nov, 1:30 PM: pdf     [ NBA salaries: JMP data file ]
• Problem set 08, due Mon 27 Nov, 1:30 PM: pdf     [ originally due Mon 20 Nov ]
• Problem set 09, due Mon 04 Dec, 1:30 PM: pdf

Lectures

• Lecture 01, Wed 06 Sep: pdf
• Lecture 02, Mon 11 Sep: pdf
• Lecture 03, Wed 13 Sep: pdf
• Lecture 04, Mon 18 Sep: pdf
• Lecture 05, Wed 20 Sep: pdf
• Lecture 06, Wed 20 Sep & Mon 25 Sep: pdf
• Lecture 07, Wed 27 Sep: pdf
• Lecture 08, Mon 02 Oct: pdf
• Lecture 09, Wed 04 Oct: pdf
• Lecture 10, Mon 09 Oct and Wed 11 Oct: pdf
• Lecture 12, Mon 16 Oct and Wed 18 Oct: pdf
• Lecture 15, Mon 30 Oct: pdf
• Lecture 16, Wed 01 Nov: pdf
• Lecture 17, Mon 06 Nov: pdf
• Lecture 18, Wed 08 Nov: pdf
• Lecture 19, Mon 13 Nov: pdf
• Lecture 20, Mon 13 Nov and Wed 15 Nov: pdf
• Lecture 21, Mon 27 Nov and Wed 29 Nov: pdf
• Lecture 23, Wed 29 Nov and Mon 04 Dec: pdf
• Lecture 24, Mon 04 Dec: pdf

Additional readings

• A short review of the fundamental probability concepts underlying this course, written by Prof. D. Freedman, University of California, Berkeley: pdf