Biostatistics II : Florin Vaida, Ph.D.

Objectives: the students will conduct more advanced regression-based statistical analyses, including: simple linear regression and correlation analysis; multiple linear regression; logistic regression; Cox proportional hazards models.  Issues of model diagnostics and analysis of residuals, model comparison and model building, and strategies for univariable and multivariable analyses will be discussed.  The analyses will be conducted in SPSS.

Course Content

Topic

Content

Correlation and simple linear
regression

Correlation and simple linear regression

Inference for simple linear
regression

Model formulation and estimation in simple linear regression; examination of residuals and model assumptions; prediction and explained variation

Multiple Linear Regression

multiple linear regression: model assumptions and interpretation; diagnostics;

Model building

model comparison and model building in multiple linear regression; polynomial regression; ANCOVA and ANOVA as multiple linear regression

Regression for proportions

Logistic regression for binary outcomes; model interpretation; comparing several groups

Logistic regression

Multiple logistic regression; model building and model comparison

Survival analysis

Kaplan-Meier curves and log-rank test

Cox proportional hazards
model

Regression methods for survival data: Cox proportional hazards model

2016 Syllabus