## STAT 453/553: Biostatistics

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.

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: