Course Detail
Units:
3.0
Course Components:
Lecture
Enrollment Information
Enrollment Requirement:
Prerequisites: PhD student in Population Health Sciences OR Department Consent
Description
This is an applied course, not mathematical. The course covers the use of statistical modeling of health and medical data, including: linear regression, logistic regression, exact logistic regression, Cox regression, Poisson regression, fractional polynomial regression, generalized linear model, and basic causal modeling (propensity score analysis). Related topics include: data and cubic spline transformations, Monte Carlo and bootstrap simulation, missing data imputation, outliers, partial and semi-partial correlation, power analysis and sample size determination, noninferiority testing, multicausality, and multiple comparisons. Topics are illustrated with the Stata statistical software. An optional Stata bootcamp is offered the week preceding the first day of class. Credit cannot be given to students who have already taken PBHLT 7100.