Applied Regression Analysis

Regression modeling is the standard method for analysis of continuous response data. This course provides theoretical and practical training in statistical modeling with particular emphasis on linear and multiple regression.

About The Course

Statistical modeling is a fundamental element of analysis for statisticians, epidemiologists, biostatisticians and other professionals of related disciplines. People in the health sciences profession rely on regression modeling to gain insight on make decisions based on a continuous flow of response data.

Focusing on linear and multiple regression, this course will provide theoretical and practical training in statistical modeling.

This is a hands-on, applied course where students will become proficient at using computer software to analyze data drawn primarily from the fields of medicine, epidemiology and public health.

There will be many practical examples and homework exercises in this class to help you learn. If you fully apply yourself in this course and complete all of the homework, you will have the opportunity to master methods of statistical modeling when the response variable is continuous and you will become a confident user of the Stata* package for computing linear, polynomial and multiple regression.

*Access to Stata will be provided at no cost for the duration of this course.

Frequently Asked Questions

What resources will I need for this class?

The only resources you will need are time and determination to learn. We will provide you will all the course lecture notes, the computer software and the solutions to the homework assignments.

What is the most practical thing I'll learn if I take this class?

You'll learn how to build a statistical model that predicts a continuous response variable from a subset of many potential predictors. Models like this are used extensively in the modern world and appear regularly in the scientific literature. This course will provide you with the skills necessary to understand and develop your own models.

If the solutions are provided with the homework, why should I do them?

If you truly want to learn and be confident in your practical understanding of regression analysis, the key will be to work through the homework exercises. Using the solutions can help you through that process. Remember, the goal isn't to get the correct answer...but to learn and understand the process.

This course was developed in partnership with Université Paris Descartes, Sorbonne Paris Cité and Charité – Universitätsmedizin Berlin – collectively under the umbrella of the Centre Virchow-Villermé for Public Health.

The Centre Virchow-Villermé for Public Health was created by the French and German governments. It responds to the desire to pool knowledge and combine competencies in the field of Public Health. In associating Sorbonne Paris Cité and Charité – Universitätsmedizin Berlin, it is the objective of the Centre to become an important point of reference in the field of Public Health.

Recommended Background

Students interested in this course should have taken a solid introductory statistical methods course. While not absolutely necessary, familiarity with some statistical software package (such as SPSS, SAS or R) would definitely be an advantage. 

We will be making exclusive use of Stata for the course lectures and homework solutions. However, students who are proficient in other packages could use those packages in place of Stata. 

Based on past experience we strongly encourage students to use Stata for this course as it will facilitate communications with the professor, the teaching assistants, as well as other students in the course.