Syllabus
Statistics
410: Introduction to Statistical Computing and Regression
Spring 2002
|
Instructor: |
Rudy
Guerra |
|
Office: |
2099
Duncan Hall |
|
Ofc Hrs: |
3:00-4:30p.m.
TTH & appt. |
|
E-mail: |
rguerra@rice.edu |
|
Phone: |
713.348.5453 |
Required Textbooks
Applied Linear Statistical
Models,
Fourth Edition (1996) by J. Neter, M.H. Kutner, C.J. Nachtsheim, W. Wasserman.
McGraw-Hill, ISBN 0-256-11736-5.
Modern Applied Statistics
with S-Plus (1999)
by W.N. Venables and B.D. Ripley. Springer, ISBN 0-387-98825-4.
Description
This
course is an applied statistics course dealing with applications of regression
and analysis of variance (ANOVA), and more generally linear models. Both fixed
effects and random effects will be presented. Briefly, we will discuss
statistical models relating predictor variables to responses. Generally, the
response will be a continuous variable, but we will also spend several lectures
on categorical responses (e.g., binomial and Poisson). Applications will
include epidemiology, genetics, botany, engineering, psychology, sociology,
economics, and business. Both observational studies and experiments will be
covered. In this class, the term "statistical computing" in the title
refers to the use of statistical software to apply the models in data analysis.
The main statistical package we will use is S-Plus, but we will also cover
basic SAS procedures for linear models. There is no prerequisite in computing
for this course.
The
emphasis of the course will be applications. We will devote some time theory,
but only when it sheds light on the methodology. A working knowledge of
differential and integral calculus is necessary and some background in matrix
algebra will be helpful. As much as possible we will discuss real data and case
studies. If you have a dataset or case study that you think may be interest let
me know and I'll try to integrate it in the lectures or assignments.
A
class webpage will be available through my homepage at: www.stat.rice.edu/~rguerra
Course Requirements
Textbook
assignments, projects, and examinations will constitute the basis for your
grade. The relative weighting is given below. You can expect weekly textbook
assignments and data analysis projects will be given approximately every two
weeks. There will be a midterm examination,
Thursday, March 14 and a comprehensive final examination.
|
Component |
Textbook |
Projects |
Midterm |
Final |
|
Percentage |
25 |
35 |
15 |
25 |
|
Meeting |
Topic |
Readings |
|
1.
T, Jan 15 |
Intro
1: Simple Linear Regression (SLR) |
Handout |
|
2.
R, Jan 17 |
Intro
2: SLR Modeling/Estimation |
1 |
|
3.
T, Jan 22 |
SLR:
Inference & Prediction |
2.1-2.6 |
|
4.
R, Jan 24 |
SLR:
ANOVA, r2. S-Plus |
2.7-2.11 |
|
5.
T, Jan 29 |
SLR:
Diagnostics |
3.1-3.7 |
|
6.
R, Jan 31 |
SLR:
Remedial measures, Simul. Inf. |
3.8-3.11,
4 |
|
7.
T, Feb 5 |
Multiple
linear regression (MLR) |
5,
6.1-6.2 |
|
8.
R, Feb 7 |
MLR
and S-Plus |
6.3-6.9,
11.1-11.3 |
|
9.
T, Feb 12 |
MLR:
SS, multicollinearity |
7.1-7.6 |
|
10.
R, Feb 14 |
MLR
Model building: Variable selection |
8 |
|
11.
T, Feb 19 |
MLR
Model building: Diagnostics |
9 |
|
12.
R, Feb 21 |
No Class |
|
|
13.
T, Feb 26 |
MLR:
Weighted LS & Robust regression |
10.1,
10.3 |
|
14.
R, Feb 28 |
Binary
regression: Logistic model |
14.1-14.9 |
|
Spring Break Mar 4-8 |
|
|
|
15.
T, Mar 12 |
Review
and/or Catch-up |
|
|
16.
R, Mar 14 |
Midterm Exam |
|
|
17.
T, Mar 19 |
One-way
ANOVA. S-Plus |
16 |
|
18.
R, Mar 21 |
One-way
ANOVA: Simultaneous inference |
17 |
|
19.
T, Mar 26 |
One-way
ANOVA: Diagnostics |
18 |
|
Spring Recess Mar 28-29 |
|
|
|
20.
T, Apr 2 |
Two-way
ANOVA S-Plus |
19 |
|
21.
R, Apr 4 |
Two-way
ANOVA: Design issues |
20 |
|
22.
T, Apr 9 |
Tow-way
ANOVA: Design issues |
22 |
|
23.
R, Apr 11 |
Analysis
of Covariance |
25 |
|
24.
T Apr 16 |
Random
and mixed effects models |
24 |
|
25.
R, Apr 18 |
Random
and mixed effects models |
24 |
|
26.
T, Apr 23 |
Special
Topics |
|
|
27.
R Apr 25 (last class) |
Special
Topics |
|