Humboldt-Universität zu Berlin - Statistics

Advanced Regression

 

Description

Regression analysis is one of the most developed and commonly used methods in the statistical toolbox. This course gives an introduction to the vast field of regression modelling techniques that extend the classical linear regression model. The course presents the foundations of regression analysis and highlights its application, interpretation, and underlying assumptions.
 

The topics of this course include a primer on the classical linear regression model, regression models for non-normal responses, and non-parametric smoothing techniques to handle non-linear covariate effects. Data examples illustrate these methods. The lecture is accompanied by an excercise that will show how to implement these approaches using statistical software packages.

 

Course Outline

The registration in the respective Moodle course is obligatory.

  1. Introduction / The Linear Model
  2. Binary Regression Models
  3. Statistical regularization and Bayesian approaches
  4. Generalized Linear Models
  5. Categorical Regression Models
  6. Generalized Additive Models

 

Literature

  • L. Fahrmeir, T. Kneib, S. Lang, B. Marx (2013): Regression: Models, Methods and Applications, Springer-Verlag
  • G.Tutz (2011): Regression for Categorical Data, Cambridge University Press
  • Simon Wood (2017, 2nd edition): Generalized Additive Models: An Introduction with R, Chapman & Hall