BIOSTATISTICAL ANALYSIS OF EPIDEMIOLOGIC DATA III
June 27 - July 1, 2016
8:30 a.m. – noon
Course Number: 140.678.11
Discusses the application and interpretation of statistical tools useful for the analysis of Epidemiologic and medical data. Included are numerous case studies using real or realistic data to further explore the properties of at least six important biostatistical techniques. The material covered includes: non-parametric survival analysis including the Cox proportional hazard model, the analysis, application and statistical structure of disease and morality rates, analysis of time/space clustering of disease illustrated by childhood leukemia, the analysis of tables with model (loglinear models) and model-free techniques, classification accuracy with emphasis on ROC curves, graphical methods of analysis applied to regression and prediction models and a comprehensive description of the analysis of case/control data contained in 2 by 2 and 2 by tables.
Learning Objectives: Upon successfully completing this course, students will be able to
- Use semiparametric statistical techniques as applied to epidemiologic data
- Apply smoothing techniques (local linear methods, kernel methods and splines)
- Utilize the Weibull hazard model to describe mortality patterns among untreated AIDS patients
- Apply commonly-used semiparametric proportional hazards model approaches.
Grading Options: Letter Grade or Pass/Fail
Course Materials: Provided in Class