Final exam review

There will be an in-class final exam on July 27. You should be prepared to take the exam on that date --- I don't anticipate giving any Incompletes this summer.

The exam will be an open book (and notes) exam --- you can use whatever books and notes that you wish to bring with you.

I found it hard to come up with good questions/problems for an in-class open book exam. In order to make it easier for you to prepare for the exam, I've emphasized certain topics/themes, as indicated below. (Roughly, the foci are bootstrap and jackkife estimates of standard error and bias, and understanding how the bootstrap world can be used to mimic the real world.) You can bring a computer if you wish to do so, but a calculator ought to suffice. (Although I haven't finalized the exam yet, my guess is that only for part (b) of Section 5 will you be requested to provide a numerical answer.)

The exam will consist of 6 sections. Each student will choose 2 sections to eliminate from consideration, and submit answers for the other 4 sections. Below I will describe the 6 sections.

Section 1

2 to 4 True/False and/or multiple choice questions pertaining to Chapters 9 through 14.

Section 2

This section will have 3 multiple choice questions pertaining to Chapters 15 through 17. One will be about hypothesis testing, one will be about obtaining an estimate of prediction error, and one will be about something I covered in class on Tue 7/25 but is not heavily emphasized in E&T.

Section 3

Something similar to Quiz 5.

Section 4

This section will consist of two questions, for which you are to provide short answers and possibly a brief explanation for one of the answers. Having a good understanding of the boostrap estimates of bias and standard error should be helpful --- understand how (10.4) on p. 125 of E&T follows from (10.1) on p. 124 of E&T when the bootstrap world is used to mimic the real world. Similarly, understand why (6.6) on p. 47 makes sense.

Section 5

A problem dealing with the jackknife estimate of standard error or the jackknife estimate of bias. There will be two parts --- in part (b) you'll be asked to give a jackknife estimate based on a very small sample (n = 3).

Section 6

Something similar to these problems from Ch. 10 and Ch. 11, having two parts. (I'll give you some hints with the messier of the two parts. To prepare for this part of the exam, I suggest that you practice doing the problems from Ch. 10 and Ch. 11 that I provided answers for.)