ANOVA

Yoav Benjamini: Email : benja@math.tau.ac.il; Phone : 6409621; Office Hours( Schrieber Room 214): Tuesday 13:30-17; Wednesday 11-12.

Lecture Hours:

Monday 10-12 Kaplon 118, Wednesday 10-11 Shrieber 007

Communication:

This homepage will be a source of information as the course progresses. In particular a newsgroup has been set up specifically for this course. By clicking HERE, you will get directly to the newsgroup. There you will get messages and announcement from us, but maybe even more important you will be able to post your questions, requests, clarifications and both the questions and the answers will be available to everyone taking the course. If some of you can contribute an answer to a question please do so.

Books:

Statistics For Experimenters

by Box Hunter and Hunter. (BHH)

Statistical Models in S

by Chambers and Hastie.

Computing:

It is assumed you know enough about the UNIX environment in order to be able to use email, and to create edit save and print a file.
The statistical software we shall use is Splus. Tutorial material is available in the following documents (pages of which can also be printed).

The introductory lecture presented at class can be found HERE .
A homepage with short explanations on Splus can be found HERE . Suggested reading in this list is: Factors , Data Frames , Model Formula , Linear Models .
Material presented in lecture on two-way analysis in Splus .
Material presented in lecture on ANOVA in Splus .
Notes on S: A programming environment for Data Analysis and Graphics Splus by Venables and Smith.
Introductory Guide to Splus by Ripley.

Try using NETSCAPE & which allows you then to open in the now active window Splus, and working in Splus while looking at the explanations at NETSCAPE.

After a while you will find that the online interactive help is enough in order to get along with the statistical software.

Topics:

Introduction: Motivating Example ; Means Variances and the t-test: Geometric approach (For those for which this is new: Boxplot)
One-way ANOVA: Projections; Normal Model; Test of the Intersection (overall) hypothesis.
The problems of pairwise comparisons
Bonferroni, Tukey, and Scheffe. Holm for ad-hoc.
FDR.
Graphical Methods.
Checking the assumptions in the one-way normal model and remedies.
Random effects model
Two-way ANOVA and the notion of interaction.
Test of non-aditivity, diagnostic plots and remedies.
Mixed Models.
Multi-way ANOVA and higher order interactions.
ANCOVA.

Please keep on looking at this list as the course progresses. Some of these will be activated later, and by clicking on them further information, material, etc. will be available.

Requirements: