Applied Statistical Regression

Autumn semester 2019

General information

Lecturer Marcel Dettling
Assistants Corinne Emmenegger, Peter Hinz
Lectures Mon 08-10 HG E 1.2 >>
Exercises Mon 10-12 (bi-weekly) HG E 1.2 >>
Course catalogue data >>
Literature

Faraway (2005): Linear Models with R
Faraway (2006): Extending the Linear Model with R
Draper & Smith (1998): Applied Regression Analysis
Fox (2008): Applied Regression Analysis and GLMs
Montgomery et al. (2006): Introduction to Linear Regression Analysis

Course content

Abstract

This course offers a practically oriented introduction into regression modeling methods. The basic concepts and some mathematical background are included, with the emphasis lying in learning "good practice" that can be applied in every student's own projects and daily work life. A special focus will be laid in the use of the statistical software package R for regression analysis.

Objective

The students acquire advanced practical skills in linear regression analysis and are also familiar with its extensions to generalized linear modeling.

Content

The course starts with the basics of linear modeling, and then proceeds to parameter estimation, tests, confidence intervals, residual analysis, model choice, and prediction. More rarely touched but practically relevant topics that will be covered include variable transformations, multicollinearity problems and model interpretation, as well as general modeling strategies.

The last third of the course is dedicated to an introduction to generalized linear models: this includes the generalized additive model, logistic regression for binary response variables, binomial regression for grouped data and Poisson regression for count data.

Notice

The exercises, but also the classes will be based on procedures from the freely available, open-source statistical software package R, for which an introduction will be held.

Announcements

Course materials

Lecture notes

The lecture notes can be found here.

The data sets used in the lecture notes can be found here.

The slides used for the lecture will be made available on this page.

Course organisation

The following table contains a tentative outline of the course, changes might apply. Further information can be found here.

Week Topic
Week 1 (16.09.2019) No lecture
Week 2 (23.09.2019) Linear Modeling and Smoothing
Week 3 (30.09.2019) Simple Linear Regression: Fitting and Inference
Week 4 (07.10.2019) Curvilinear Models, Variable Transformations
Week 5 (14.10.2019) Multiple Linear Regression: Model and Fitting
Week 6 (21.10.2019) Multiple Linear Regression: Inference and Prediction
Week 7 (28.10.2019) Extensions: Categorical Variables, Interactions
Week 8 (04.11.2019) Model Diagnostics: Standard Residual Plots
Week 9 (11.11.2019) Model Diagnostics: Advanced Techniques
Week 10 (18.11.2019) Multicollinearity and Variable Selection
Week 11 (25.11.2019) Modeling Strategies, Cross Validation
Week 12 (02.12.2019) Generalized Additive Modeling (GAM)
Week 13 (09.12.2019) Generalized Linear Modeling (GLM)
Week 14 (16.12.2019) Grouped Data, Poisson Regression

Exercise classes

Exercises will be held roughly bi-weekly, see below. On these dates, the exercise classes will take place from 10:15 to 11:55 in HG E 1.2. The first exercise class is meant to be an opportunity for you to ask questions regarding the software R. The material you should be familiar with consists of the R tutorial and exercise sheet 1. Also further on, R will be used during the exercises so that you are expected to bring your laptop to the classes. Starting with the second exercise class, the idea is that there will be a discussion of the old exercise sheet (common problems) and a discussion of the new exercise sheet (hints and theory as needed) taking at most one hour. Afterwards, you work on the problems using the computer; the assistants will be there to give instructions and support.

  • September 23, 2019
  • October 07, 2019
  • October 21, 2019
  • November 04, 2019
  • November 18, 2019
  • December 02, 2019
  • December 16, 2019

Series and solutions

The solved exercises should be placed in the corresponding tray in HG J68 on the due date by 4pm at the latest. Only solutions to the exercises with your most important findings and answers shall be handed in, but no R script files and lengthy compilations of output or figures. It is much more important to give your understanding and interpretation of your findings than it is to provide many figures and numbers.

Help with R

During the first exercise class you will have the opportunity to ask questions regarding the software R. Further material can be found following the links below.

R homepage
R studio homepage
Getting help with R
Online R course (in German)
Try R