Last
Update: January 9, 2004.
Time &
Location: MWF
9:00-9:50 pm, in Keck Hall room 107
Instructor: Kenneth R.
Hess, Ph.D.
Office: MD
Anderson Faculty Center, FC2.3030
Phone: 713-794-4157
Email: khess@odin.mdacc.tmc.edu
Main Text:
Woodward M. Epidemiology: Study Design and Data Analysis. Boca Raton:
Chapman & Hall / CRC, 1999.
Supplemental Texts: Venables WN, Ripley BD. Modern
Applied Statistics with S. 4th ed. Springer, 2002; B Everitt, S
Rabe-Hesketh, Analyzing Medical Data
Using S-Plus. Springer, 2001
Teaching
Assistants:
Office:
Phone:
Email:
Course
Description: This course covers
the design of biomedical and epidemiological studies and the analysis of the
resulting data. Because this is primarily a course for statistics majors, the
applied methods will be related to theory whenever practical. Emphasis
will be placed on the similarity between various forms of analysis and
reporting results in terms of measures of effect or association. Emphasis
will also be given to identifying statistical assumptions and performing
analyses to verify these assumptions. S-Plus (R) will serve as the basic
computing software.
Course
Prerequisites: STAT 410
Computing: Weekly homework will involve statistical
analyses that will often require access to sophisticated statistical software.
In general, students may use the software of their choice. However, a few
assignments may require S-PLUS, and computing instructions, when given, will be
in S-PLUS.
Assignments:
Weekly homework will be a mix of
data analysis, computer simulations, report writing, and answering questions
relating to relevant statistical theory and methods. Students are
encouraged to seek help from the TA, theinstructor or other students as to the
methods of solution, but the submitted report should reflect only the students'
work. Reports will be written legibly (or typed), and be grammatically
correct.
Grading:
60
% Homework
30
% Final
10
% Oral Participation
Syllabus:
Wk Topic
1 Overview, study design, data, graphs,
inference
2 Comparing means (t-tests, ANOVA,
rank-based)
3 Simulations, permutation tests,
bootstrapping,
Comparing
proportions (binomial, chi-square test)
4 Survival analysis (censoring, hazard
function) 5 Regression analysis (linear, logistic,
residuals)
6 Proportional hazard regression
analysis
7 Rates and counts (standardization,
Poisson regression)
8 Multivariable analysis (stratification,
regression)
9 Model assessment (goodness of fit,
predictive accuracy)
10 Biomarkers, replication, multilevel data
(pairing)
11 Longitudinal data, multilevel models,
multistate models
12 Supervised and unsupervised data mining
13 Microarray data analysis, diagnostic
tests
14 Power and Sample Size, Clinical trials
S-Plus
Functions:
Datasets:
Homework
2: Scottish Heart Health Study
US
Age Distribution for 1940 Given as Relative Frequencies
Misc.
Links:
List of
Errata from Woodward's Text
Any
student with a disability requiring accommodations in this course is encouraged
to contact the professor after class or during office hours.
Additionally, students should contact Disability Support Services in the Ley
Student Center.
Send problems or suggestions to shuhan@rice.edu