STATS 330 Course Information 2014

Lecturers:
Steffen Klaere
Department of Statistics
Room 219, Building 303
Telephone : 373 7599 extension 85237 Fax: 373 7018
email: s.klaere@auckland.ac.nz

Arden Miller
Department of Statistics
Room 229C, Building 303
Telephone : 373 7599 extension 85053 Fax: 373 7018
email: a.miller@ auckland.ac.nz

Office Hours:
Office hours are:

Steffen: 9:30 - 10:30 Thursday and Friday.

Arden: Wednesday 9-10 and Thursday 12-1

Students may expect to find us in our offices and available for consultation
during these times. Outside office hours we don't guarantee to be in, but
welcome enquiries if we are. Alternatively, make an appointment with us.

Lectures:
Monday 3pm Clock Tower 039, Wednesday 8am MLT1, and Thursday 11am
in MLT1. First class meeting is on Monday 20th July. The first half of the
course will be taught by Steffen Klaere, the second half by Arden Miller.

Tutorials:
Every week on Fridays we have three hour-long tutorial sessions, from 11-
12, 3-4 and 4-5.They are held in the ground floor tutorial laboratory in
Building 303S, Room 303S-G75. We operate these as drop-in sessions, so
you can come at anytime during these three hours. Usually a worksheet is
available for you to work through, so you can develop the R skills required
for the current assignment. Help is also available for any aspect of the
course. NB: Tutorials begin in the first week.

Course Content:
This course provides an introduction to the process and procedures of
statistical modelling. The topics to be covered include graphical methods,
multiple regression, regression diagnostics, analysis of variance and analysis
of covariance. We also consider some extensions of this kind of analysis to
generalized linear models, including log-linear models and logistic regression
models, with particular emphasis on the analysis of contingency tables.

Learning Outcomes:
At the conclusion of the course, you should have be able to
Explore data graphically,
Make a sensible choice of model, based on the data, and the scientific question
being addressed
Fit the model using R
Critically examine the model fit, and make adjustments as necessary
Draw sensible conclusions from the analysis
Communicate these conclusions to a lay audience.

Computing:
To do the assignments you will need to use a computer. You can either use
one of the University computer laboratories, or your own personal
computer. Some help on computing issues is available in the large computer
laboratory in the basement of the Building 303S.

The computer language used in the course is R. If you are using your own
computer, you will need to load R onto it. See the course website for
instructions.

Assignments:
For students enrolled in STATS 330, will be five assignments. The due dates
are given in the Course Planner below. The assignments will typically call for
a computer analysis of a set of data. These must be typed, using
RMarkdown, Word, or Latex.

Test:
Instead of a lecture, there will be a test of one hour's duration on Thu Sept
17, at the usual lecture time in two separate lecture theatres. The test will
be "closed book".

Examination:
The final examination for both STATS 330 and STATS 762 will be held at a
time and place to be arranged. It will also be "closed book", and be of 3
hours duration. The exam will be partly multiple-choice.

Texts:
The course book for this course is available on the class web page. In
addition, electronic copies of all the lecture slides (with voice-over) are
available on the class web page. A reading list is also given below.
Web Page:
All the course materials are available on the Web. Follow the link on the
class Cecil page. All assignments will be distributed via the Web and via
CECIL. There is also a bulletin board, which you should consult regularly.
You can also access the course page via the URL
https://www.stat.auckland.ac.nz/~stats330/

Assessment:
The final mark for the year is calculated on the basis of the assignments,
the test and the end of year examination. The assessment components for
STATS 330 are valued as follows (total 100%)

Assignments: 20%
Test 20%
Examination 60%

In order to pass the paper you must get 50% out of the total of 100%.
Note: It is very important that you attempt ALL of the assignments and sit
the test. Assignments are an essential part of this course as they give you
practice in applying the theory and techniques presented in lectures to
actual problems. You will find it difficult to master the ideas discussed in the
course without the practice you get from doing the assignments.

Collaboration:
It is our view that discussion with other students is an important part of the
learning process and we encourage you to discuss problems with each other
(and us!) However, you must not copy the details of another person's
assignment. In other words, you can work together to decide how to do an
assignment, but you must write up your own solutions. You must not
collaborate during tests and examinations.

Reading List:
We have found the following books useful in the preparation of the course.
Some of them are classic works - most of the material in this course is very
traditional, apart from the use of R.

J Adler (2010). R in a Nutshell. OReilly.

A Agresti, (2002). Categorical Data Analysis, 2nd Ed, Wiley.

JM Chambers, WS Cleveland, B Kleiner and PA Tukey, (1983). Graphical

Methods for Data Analysis, Duxbury Press.

JM Chambers and TJ Hastie, (1992). Statistical Models in S, Wadsworth.

S Chatterjee, AS Hadi (2006). Regression Analysis by Example (4th Ed),

Wiley.
WS Cleveland, (1994). The Elements of Graphing Data (revised Ed), Hobart
Press.

WS Cleveland, (1993). Visualizing Data, Hobart Press.

RD Cook and S Weisberg, (1982). Residuals and Influence in Regression,

Chapman and Hall.

RD Cook and S Weisberg, (1999). Applied Regression Including Computing

and Graphics, Wiley.

P Dalgaard, (2002). Introductory Statistics with R, Springer. AJ Dobson,

(2002). An Introduction to Generalized Linear Models (2nd Ed), Chapman &
Hall.
NR Draper and H Smith, (1998). Applied Regression Analysis (3rd Ed),
Wiley.

B Efron and RJ Tibshirani (1993). An Introduction to the Bootstrap.

Chapman and Hall, London.

J Fox, (1997). Applied Regression Analysis, Linear Models, and Related

Methods, Sage Publications.

J Fox, (2002). An R and S-Plus Companion to Applied Regression, Sage

Publications.

FE Harrell (2001). Regression Modeling Strategies. Springer, New York.

T Hastie and RJ. Tibshirani, (1990). Generalized Additive Models. Chapman

and Hall.

T Hastie, R Tibshirani and J Friedman. (2009). The Elements of

Statistical Learning : Data Mining, Inference, and Prediction (2nd ed).
Springer.

DW Hosmer and S Lemeshow, (2000). Applied Logistic Regression (2nd Ed),

Wiley.

DG Kleinbaum and M Klein, (2002). Logistic Regression : a Self-Learning

Text. New York: Springer.

S Menard, (2002). Applied Logistic Regression Analysis. Thousand Oaks,

Calif.: Sage Publications.
DC Montgomery, EA. Peck and GG Vining. (2001). Introduction to Linear
Regression Analysis (3rd Ed), Wiley.

P Murrell (2011). R Graphics, 2nd ed. Chapman and Hall.

P Murrell (2009). Introduction to Data Technologies. Chapman and Hall.

WN Venables and BD Ripley, (2004). Modern Applied Statistics with S,

4th Ed, Springer.

WN Venables and DM Smith, (2002). Introduction to R, Springer.

S Weisberg, (1985). Applied Linear Regression (2nd Ed), Wiley.

E de Jonge and M van der Loo (2013). An introduction to data cleaning with
R, Discussion Paper 13, Statistics Netherlands.
 Course Planner: Names on lectures are tentative and subject to changes

Week Starting Monday Wednesday Thursday Friday

Lecture 1. Lecture 2. Lecture 3. Tutorial 0.

1 20/07/2015 Introduction. Data Cleaning. Graphics.

Lecture 4. Lecture 5. Lecture 6. Tutorial 1.

2 27/07/2015
Trellis Graphics. Multiple Regression. Inference.

Lecture 7. Lecture 8. Lecture 9. Tutorial 2.

3 03/08/2015
Prediction. Collinearity. Diagnostics 1.

Lecture 10. Lecture 11. Lecture 12. Tutorial 3.

4 10/08/2015 Diagnostics 2. Diagnostics 3 Diagnostics 4.
Assignment 1 due.

Lecture 13. Lecture 14. Lecture 15. Tutorial 4.

5 17/08/2015
Variable Selection 1. Variable Selection 2. Case Study.

Lecture 16. Lecture 17. Lecture 18. Tutorial 5.

6 24/08/2015 Using Factors. ANOVA models. Data Science.
Assignment 2 due.

Mid-Semester Break

7 14/09/2015 Lecture 19. Lecture 20. Midterm test. Tutorial 6.

Large scale models. Categorical Response. in-class.
Assignment 3 due.

8 21/09/2015 Lecture 21. Lecture 22. Lecture 23. Tutorial 7.

Multiple Logistic Multiple Logistic Multiple Logistic
Regression 1. Regression 2. Regression 3.

9 28/09/2015 Lecture 24. Lecture 25. Lecture 26. Tutorial 8.

Multiple Logistic Prediction in Logistic Binary ANOVA.
Regression 4. Regression.

10 5/10/2015 Lecture 27. Lecture 28 Lecture 29. Tutorial 9.

Poisson Regression. Contingency Tables. Two-dimensional
Assignment 4 due. Contingency Tables.

11 12/10/2015 Lecture 30. Lecture 31. Lecture 32. Tutorial 10.

Higher-dimensional Simpson's Paradox Goodness of Fit.
Contingency Tables. and Association
Reversal.

12 19/10/2015 Lecture 33. Lecture 34. No Lecture. No Tutorial

Final Topics. Revision.
Assignment 5 due