STAT 6338

Advanced Statistical Methods, Part II

Spring 2011
Time: MW 700 – 815 pm
Room: F0 2.608
Instructor: Dr. Michael Baron
Office: FO 2.602-E
Phone: 972-UTD-6874
Internet: http://www.utdallas.edu/∼mbaron/6338
http://https://elearning.utdallas.edu/ (for grades, chat, discussion)
Office hours: Monday, Wednesday 425 – 525 pm
Texts: (1) Applied Linear Statistical Models by Kutner, Nachtsheim, Neter, and Li,
5-th edition, McGraw-Hill, 2004 (required): Parts 4–6
(2) SAS and SPSS Program Solutions for Use With Applied Linear Statistical
Models by W. D. Johnson and W. H. Replogle (not required)
or
A Step-by-Step Approach to Using the SAS System
for Univariate and Multivariate Statistics
by L. Hatcher and E. Stepanski, SAS Publishing (not required)
(3) SAS OnlineDoc(R) , Version 8
at http://www.okstate.edu/sas/v8/sashtml/main.htm

Grading: Homework assignments = 10%

SAS projects = 20%
Two midterm exams = 20% each
Final exam = 30%

97 − 100 % = A+ 90 − 97 % = A 86 23 − 90 % = A−
83 13 − 86 23 % = B+ 80 − 83 31 % = B 75 − 80 % = B−
70 − 75 % = C+ 60 − 70 % = C 0 − 60 % = F

Homework: Homework will be assigned weekly and graded quickly. A steady effort to work out
all the assigned problems is essential for learning statistical methods and successful
performance in this course. Brief or full homework solutions will usually be given.

Exams: Each exam will cover the material learned during the previous month. During an exam,
you may use your notes, textbook, calculator, and tables of distributions.

Projects: Every 3-4 weeks, a project will be assigned, where recently acquired statistical methods
will be used to analyze various data sets - the ones given on a CD attached to the textook
as well as some others. Projects should be done in SAS. A report containing the SAS
code, only essential parts of the output, your comments, results, and answers should
be submitted for grading.
 Schedule
Week Topics Chapters

1-5 Categorical predictors. Experimental Design, main concepts.

One-way ANOVA. Multiple comparisons. ANOVA diagnostics. 15-18
PROC ANOVA, GLM, NPAR1WAY.
Midterm I
6-8 Two-way ANOVA. Interactions. Testing contrasts. 19-20
9 Higher order treatment structures. 24
10-12 Balanced designs. Expected mean squares. Estimation of variance components.
PROC VARCOMP. Randomized block designs. Random effects. 21, 25
Midterm II
13 Latin squares, graeco-latin squares, and similar designs. 28
14 Nested designs. PROC NESTED. Repeated measures designs. 26-27
15 Split-plot designs, split-split-plot designs, strip-plot designs. 27
16 Other topics and preparation for the Final Exam.
17 Final Exam

Where is SAS:
• At UTD on Apache (free)
– directly
– remotely from home with one text window (use Putty)
– remotely from home with many windows (use Xming + Putty)
• At UTD Windows computer labs (free but check availability of SAS)
• SAS OnDemand on SAS server until May 31, 2011 (free, look for the invitational email
this week)
• Learning Edition of SAS ($199 until Dec 31, 2011)
• Full SAS (costly)