This course is a continuation of STAT 341, where you first began studying mathematical statistics: the mathematical theory behind the basic statistical techniques used today. Time permitting, we will look at estimation, confidence intervals, hypothesis testing, and regression. Roughly, this equates to Chapters 5-9 and 11 of the textbook.

The official prerequisite of this course is STAT 341, and I will assume that you are familiar and comfortable with the material covered in that course. We will also make use of quite a bit of calculus and matrix algebra.

**E-mail:** hanna[at]stat.washington.edu

Please include "[342]" (using square brackets) in the subjects of
your e-mails; this'll help me separate if from my other, mostly junk,
mail. Also, please send only plain text messages, no html.

**Office Hours with HJ:** Every Monday and Friday, in Padelford
B-220, from
10-11am. Office hours will begin Friday, March 30th.

If you cannot make it to my office hours, please schedule an
appointment. All appointments will take place in my office, Padelford
B-220.

**TA:** The teaching assistant for this course is Roopesh Ranjan.
His email address is roopesh.ranjan[at]gmail.com, and office phone
206-543-8471.

**Office hours with RR:** Tuesdays 3-4pm in Padelford B-314, and
Thursdays 1-3 pm in library of McCarty Hall (where the Stats Help
Centre is).
The office hours begin the second week of classes.

**Lectures** will be held in THO 134/135, as scheduled.

**Tutorials** will take place every Wednesday in THO 134, from 11:30
to 12:20. The tutorials will start April 4th. On March 28th you will
have a regular lecture.

**Tutor & Study Centre** Some of RR's office hours will take place
in the Statistics Tutor & Study Centre (see above for dates/times).
However, you are welcome to take advantage of the resource on your own as
well. Click here
for more info.

There is a copy of this text available (soon) for
short-term loan from the Mathematics Research Library in
Padelford Hall C-306. I have also requested that * Mathematical
Statistics and Data Analysis * by John Rice be placed on reserve. The
book offers nice explanations of the material we will be covering.

I have created a message board for this course. It is available here. Information on how to use the message board can be found here.

The idea of the message board is for you to be able to communicate with HJ/RR/other students quickly and easily, regarding any topic related to the course. For example, if you're stuck on a question and need a hint, you could post your question here. Both HJ and RR will check the message board regularly. Please, do not post complete solutions to any question here!

The natural disclaimer is that I have not used this technology in a course yet, and I expect that some ironing out of wrinkles will be necessary before we are all comfortable with how to use the message board. Hopefully it will be a success.

There will be roughly 6 assignments, one mid-term test and a final in this course. The assignments will be posted below, and will be worth 30% of your final grade.

Note: all assignments are due at the **beginning** of each lecture
on the day that they are assigned. **There are no exceptions. I do not
accept late assignments.**

Assignment 1. Due: Friday, April 6th.

Assignment 2. Due: Wednesday, April 18th.

Assignment 3. Due: Monday, April 30th.

Assignment 4, with problem and data set. Due: Friday, May 18th.

Here is the solution to Q1 on Assignment 4.

Assignment 5 with data. Due: Wednesday, May 30th.

Bonus Assignment Due: see details.

**Solutions** to your assignments are available here.

Midterm: Friday, May 4th, 30%

Final: June 6th, 2:30-4:20pm 40%

- The tests are all closed book tests. You will probably need a calculator - only non-programmable calculators are allowed.

- There will be no make-up tests scheduled. In case of illness supported by medical documentation, your grade will be based on your other grades, pro-rated accordingly.

- Course work will be handed back during lectures. If you
disagree with the mark you have received, you should submit a
request for regrading in writing to the instructor within
*one week*of when the work was returned.

- You are free to discuss your homework assignments with others, in fact, I would encourage you to do so. However, you must write up your own solutions. I cannot stress this last part enough: not only is the right thing to do, but it's also how you learn the material. The assignments are due at the beginning of the class on the due dates. I do not accept late assignments.

- Attendance at all lectures is very important. In particular, some of my notation may differ from what the textbook does, and you will be expected to be familiar with my notation.

- The course is a mathematical statistics course, and the best way to learn mathematics is to do mathematics. Make sure you keep up with the non-credit practice problems. Come see me (HJ) or RR if you have troubles. Understanding the lectures and working through the assignments and practice problems is the key to your success in this course.

- If you have any concerns/suggestions etc., please let me (or Roopesh) know. This is what we're here for.

NB. In square brackets I am also putting reference chapters for you. LM stands for the course textbook, and R for the Rice reference available in the library.

We have covered the following sections/topics:

** Week 1 : **

** Week 2 : **

** Week 3: **

** Week 4:**

**Week 5:**

The two-sample t-test and associated CI [ML 9.2,9.5; R 11.2]

Two sample t-test con't. Test for equal variance [ML 9.3]; Test for
two
bernoulli parameters and CI [ML 9.4, 9.5]

END OF COVERAGE FOR MIDTERM.

**Week 6:**

Review and Midterm!

**Week 7:**

Paired t-test & Wilcoxon Signed Rank test [ML 13.3, 14.3, R 11.3]

Wilcoxon Signed Rank test con't. Intro to Regression (LSE) [ML 11, R
13]

**Week 8:**

Jen's visit. Regression con't. (mean, var, covar of LSE)

More on regression: normal assumption, testing the betas, R^2,
interpreting computer output, assessing fit
(perhaps not in that order)

**Week 9:**

Finishing up Regression (MLEs, "centering").

Start Anova (Boxplots, Anova table, Tukey's multiple comaprisons)
[ML 12.1-12.3, R 12.1, 12.2]

**Week 10:**

Holiday!

Finish up Anova: Tukey and Bonferroni multiple comparisons [R 12.2]

Review and summary.

**Practice** (non-credit - do not hand these
in) **problems**:

**Week 1:**

5.2.3, 5.2.4, 5.2.11, 5.2.19

5.4.6, 5.4.11, 5.4.19

5.5.2

5.6.1

5.7.2, 5.7.3a, 5.7.4

5.3.9, 5.3.21

**Week 2:**

7.3.2, 7.3.4, 7.3.5

7.4.15, 7.4.9a

7.5.9 (but you can do variance, not st.dev.)

6.2.1, 6.2.2, 6.4.3, 6.2.8

**Week 3:**

Handout.

6.5.1, 6.5.2, 6.3.1

**Week 4:**

7.4.17, 7.5.8, 7.5.13

Handout.

**Week 5:**

9.2.1, 9.2.5; for each question though, calculate the p-value; you
could also try doing side-by-side boxplots.

9.5.1, 9.5.2

9.3.3, 9.3.5

9.4.1, 9.4.3

9.5.11

**Week 6:**

STUDY!

**Week 7:**

13.3.1, 13.3.3

14.3.3, 14.3.5 (Wilcoxon only) NB. You should use the Wilcoxon tables
here, and not the normal approximation. Hence, your answers won't match
those of the text.

11.2.4, 11.2.5, 11.2.6, 11.2.9, 11.2.12

**Week 8:**

11.2.27, 11.3.12, 11.3.8, 11.3.2

11.4.16

More practice problems, with
plot 1 and plot 2.

**Week 9/10:**

12.2.2, 12.2.3 (for each question, draw side-by-side boxplots, and
find the p-value, regardless of what the question asks)

12.2.7

12.2.12, 12.3.4, 12.3.6

Repeat the above Tukey problems, but use the Bonferroni method
instead. Is the answer the same or different?

More practice problems, with plots, plots, and data, data.