University of Konstanz
Algorithmik
Prof. Dr. Ulrik Brandes

Generalized Linear Models (Summer 2017)

+++ News +++

This course covers the basic theory, methodology, and application of generalized linear models (GLMs). GLMs are a class of models for the analysis of quantitative and qualitative data. This class subsumes linear models for quantitative responses (multiple regression, analysis of variance and analysis of covariance), binomial models for binary response (logistic regression and probit models), log-linear models for categorical data, Poisson models for count data and survival models for failure time data as special cases. In the lecture on Wednesday the mathematical aspects of the different models are explained, while in the tutorial on Thursday the application of these models is treated using the R software. Case studies drawn from social, economic, engineering, and behavioral sciences are used to illustrate the estimation, assessment and interpretation of GLMs.

Prerequisites Good knowledge of basic mathematical and statistical concepts (e.g. definition of probability and random variables). Strong mathematical soft skills (e.g. the ability to understand and work with mathematical definitions and theorems, elements of linear algebra)

Schedule

Lecture (V. Amati) Wed 08:15-09:45 in G 306
Tutorial Thu 17:00-18:30 in Z 613
Exams (written) 02.08.2017, 10.00, P602; 11.10.2017, 10.00, P602

Exercises

Most documents are only locally accessible - see possibilities for remote access.

-->
no. online due tutorial download data solution
0 26 April 2017 03 May 2017 04 May 2017 No assignment - -
1 03 May 2017 10 May 2017 11 May 2017 Sheet 1 - Solution 1
2 10 May 2017 17 May 2017 18 May 2017 Sheet 2 rent Solution 2
3 17 May 2017 24 May 2017 01 June 2017 Sheet 3 car Solution 3
4 24 May 2017 31 May 2017 01 June 2017 Sheet 4 bmi Solution 4
5 31 May 2017 07 June 2017 08 June 2017 Sheet 5 pima Solution 5
6 07 June 2017 14 June 2017 22 June 2017 Sheet 6 pima Solution 6
7 14 June 2017 21 June 2017 22 June 2017 Sheet 7 rent_munich , advertising Solution 7
8 21 June 2017 28 June 2017 29 June 2017 Sheet 8 usedCars Solution 8
9 28 June 2017 05 July 2017 06 July 2017 Sheet 9 pima Solution 9
10 05 July 2017 - 13 July 2017 Sheet 10 - Solution 10
11 12 July 2017 19 July 2017 20 July 2017 Sheet 11 pima Solution 11
12 19 July 2017 26 July 2017 27 July 2017 Sheet 12 - Solution 12

Material

Most documents are only locally accessible - see possibilities for remote access.

Slides and lecture notes

Tutorial

Software

Literature

The course content is covered in lecture notes that are made available. Further reading:

General text books for GLM

Linear regression model Binary logistic regression Multinomial logistic regression For an introduction to statistics and probability theory:

Further information