Applied Statistical Regression
Autumn semester 2019
General information
Lecturer  Marcel Dettling 

Assistants  Corinne Emmenegger, Peter Hinz 
Lectures  Mon 0810 HG E 1.2 >> 
Exercises  Mon 1012 (biweekly) HG E 1.2 >> 
Course catalogue data  >> 
Literature 
Faraway (2005): Linear Models with R 
Course content
AbstractThis 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.
ObjectiveThe students acquire advanced practical skills in linear regression analysis and are also familiar with its extensions to generalized linear modeling.
ContentThe 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.
NoticeThe exercises, but also the classes will be based on procedures from the freely available, opensource statistical software package R, for which an introduction will be held.
Announcements

Past exams:
2018 Summer
2018 Winter
2017 Summer
2017 Winter 
August 21st, 2019:
Beginning of lecture and exercise class: Monday, 23.09.2019.
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 biweekly, 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.
Exercises  Solutions  Due date 

Series 1 (Update: 25.09.2019)  Solution 1  NA 
Series 2  Solution 2  September 30, 2019 
Series 3  Solution 3  October 14, 2019 
Series 4  Solution 4  October 28, 2019 
Series 5  Solution 5  November 11, 2019 
Series 6  Solution 6  November 25, 2019 
Series 7  Solution 7  December 09, 2019 
Series 8  Solution 8  NA 
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