PS 532 --- Winter 2007

This is the website for PS 532, which is a course covering the linear model and some extensions for graduate students in the Political Science department of the University of Illinois @ Urbana-Champaign.

Syllabus here.

Announcements

Examples and Handouts that I've worked out including Sweave files to recreate the analyses. The PDF files in this directory show all of the results of the analysis. The files that end with .Rnw include both R code and LaTeX source to create the .pdf file (after running Sweave("filename.Rnw") in R, and then processing the resulting .tex file with pdflatex. Right now the .txt files are datafiles.

Readings that are not online or in the required textbooks are available for you to copy in the usual place in 361 Lincoln Hall, or via the previous link, or on Compass .

Assignments .

top

Handouts

You can find the handouts (both the sources and the pdf versions) here .
top
top

Links on Scientific Computing

Learning to Compute and Collaborate Effectively

Gregory Wilson has a great online course on this topic.

Learning R

Learning LaTeX

top
top

Readings and Supplementary Materials

Resampling and the Bootstrap

Bob Stine's lectures for his mini-course on the bootstrap are here

A great book on the topic, Bootstrap Methods and Their Application by A.C. Davison and D.V. Hinkley.

And, of course, An Introduction to the Bootstrap by Efron and Tibshirani.

Some articles by Bradley Efron

Computers and the Theory of Statistics: Thinking the Unthinkable. Bradley Efron. SIAM Review, Vol. 21, No. 4. (Oct., 1979), pp. 460-480.
Abstract
This is a survey article concerning recent advances in certain areas of statistical theory, written for a mathematical audience with no background in statistics. The topics are chosen to illustrate a special point: how the advent of the high-speed computer has affected the development of statistical theory. The topics discussed include nonparametric methods, the jackknife, the bootstrap, cross-validation, error-rate estimation in discriminant analysis, robust estimation, the influence function, censored data, the EM algorithm, and Cox's likelihood function. The exposition is mainly by example, with only a little offered in the way of theoretical development.

A Leisurely Look at the Bootstrap, the Jackknife, and Cross-Validation. Bradley Efron; Gail Gong. The American Statistician, Vol. 37, No. 1. (Feb., 1983), pp. 36-48.
Abstract
This is an invited expository article for The American Statistician. It reviews the nonparametric estimation of statistical error, mainly the bias and standard error of an estimator, or the error rate of a prediction rule. The presentation is written at a relaxed mathematical level, omitting most proofs, regularity conditions, and technical details.

Interpretations of Probability

Controversies in the Foundations of Statistics Bradley Efron. The American Mathematical Monthly, Vol. 85, No. 4. (Apr., 1978), pp. 231-246.

Why Isn't Everyone a Bayesian? B. Efron The American Statistician, Vol. 40, No. 1. (Feb., 1986), pp. 1-5.
Abstract
Originally a talk delivered at a conference on Bayesian statistics, this article attempts to answer the following question: why is most scientific data analysis carried out in a non-Bayesian framework? The argument consists mainly of some practical examples of data analysis, in which the Bayesian approach is difficult but Fisherian/frequentist solutions are relatively easy. There is a brief discussion of objectivity in statistical analyses and of the difficulties of achieving objectivity within a Bayesian framework. The article ends with a list of practical advantages of Fisherian/frequentist methods, which so far seem to have outweighed the philosophical superiority of Bayesianism.

top