University of Minnesota, Twin Cities     School of Statistics     Stat 5601     Rweb

# Statistics 5601, Fall 2001, Prof. Geyer

## The Final Exam

The final exam (take-home) is posted. You may start work now, if you like. Follow the instructions and do your own work with no discussion with anyone except the instructor.

It is due at the time the regularly scheduled exam would finish: 6:00pm, Friday December 21. You can hand it in that late, but I expect everyone to be done long before.

New! It wasn't clearly stated that all confidence intervals were to be 95% (or as close to 95% as possible without being less than 95%). The exam has been edited to make this clear.

## Announcements

New! Homework solution #6 posted.

New! Section on material covered updated and finished, now covers the material for the final exam.

Web page on bootstrap hypothesis tests simplified as discussed in class. In particular in the section on inverting bootstrap percentile confidence intervals the computation of the lower-tailed P-value has been much simplified and the accompanying comments changed to match.

Error on second example of subsampling bootstrap confidence intervals fixed. Confidence interval now a lot smaller.

Grade distribution for second midterm posted.

## Exam Dates

 First Midterm Wednesday, October 24 Second Midterm Take-home: handed out Monday, November 19, due Wednesday, November 21. Final Exam Take-home: handed out December 12, due 6:00pm, Friday December 21, but can be handed in any time before.

## Material Covered

### First Midterm

Generally the material for the first two homeworks.

In particular, Sections 3.1, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.11, 4.1, 4.2, 4.3, 4.5. 5.4, and 11.5 in Hollander and Wolfe and the related web pages

### Second Midterm

Generally the material for homeworks three through five.

In particular, Sections 6.1, 6.2, 7.1, 8.1, 8.2, 8.3, 8.5, 11.1, and 11.6 in Hollander and Wolfe and the related web pages

and Chapters 6, 7, 8 (sort of, see below), and 9 in Efron and Tibshirani and the related web pages

Note that we didn't actually do most of Chapter 8 in Efron and Tibshirani. In fact we didn't cover any of the methods they recommend there. Instead we covered the subsampling bootstrap (Politis and Romano, 1994), which is a competitor of the methods Efron and Tibshirani cover.

### Final Exam

Final exam is cumulative, so to start, it covers everything above that was on the midterms.

Since the second midterm, we've added Chapters 12, 13, and 14 of Efron and Tibshirani and the related web pages

(actually, we also covered Chapter 10 of Efron and Tibshirani and there was a related web page on bias estimation, but we didn't do any homework on it so it won't be on the exam).