Statistics 618 (Fall 2010)


Instructor: David Pollard
When: Tuesday, Thursday 10:30 - 11:45 (might be moved)
Where: 24 Hillhouse, basement classroom
Office hours: tba
TA: David Pollard
Problem session: to be arranged if needed
Other: courses taught by DP in previous years
Short description: A careful study of some standard asymptotic techniques in statistics and econometrics, and their modern refinements. Topics selected from classical likelihood theory and M-estimation; empirical process methods; concentration inequalities; semiparametric models; local asymptotic normality; concepts of efficiency. Prerequisites: knowledge of probability at the level of STAT 600b.
Intended audience: Students (both graduate and undergraduate) who have had some exposure to measure theoretic probability and who need to understand some asymptotic theory at the reserch level.
Text: For many years I have been working on a book [Asymptopia: current table of contents] on modern asymptotic theory. Recently I decided it is getting too big; I decided to split a big manuscript into two or three smaller books. Some of the material will be covered in Stat 618.
Chapters from Asymptopia will be posted in the Handouts directory after they receive another edit.
Grading: The grading method will depend on the the audience for the course. I am leaning towards a variation on the method I used for Stat 603 last year, but with more traditional homework sheets.
Topics: Tentative list. The actual material covered will depend, in part, on the backgrounds of students in the class.
  • Most of the current Chapters 1 through 4.
  • Something on contiguity (Chap 7).
  • Something on Hellinger differentiability (Chap 10) and Local Asymptotic Normality (Chap 11).
  • At least one of the modern ways to think about efficiency.
  • If time permits, some introduction to ''empirical process methods" (Part III).
  • I would dearly love to include some material on "Le Cam theory", maybe along the lines of what I did in my Beijing Lectures this past summer. Unfortunately, on top of the other possible topics, this material might be more than can reasonably be packed into one course.

DBP 1 Sept 2010