Multivariate Analysis

Responsible Faculty: School of Biology (FBM-BIO)

Lecturer: Zoltán Kutalik
Teaching Assistant(s): Sina Rueger, Aurelien Mace


Term: Autumn semester (10 hours)
Teaching language: English
Credits: 2.00


The goal of this course is to introduce the basic elements of a multivariate analysis. A generalization of several basic concepts of statistics to the multivariate context, where many variables are collected per individual, will thus be provided, which will also consolidate the statistical competences of the students. Particular attention will be paid to the interpretation of the results in scientific papers which have used these techniques. During the exercises, the students will learn how to perform the necessary calculations of a multivariate analysis in an autonomous way, using the free statistical software R (which has already been used during the exercises of the statistical course taught during the second year).


1. General presentation of a multivariate analysis; recall of basic elements of linear algebra; comparison of vectors of means between two populations.

2. Principal component analysis: how to reduce the dimensionality of the data with a minimal information loss.

3. Discriminant analysis: how to explain and to predict the membership of individuals to pre-specified groups. Multivariate analysis of variance: comparison of vectors of means between m>2 populations.

4. Clustering : how to optimally define groups of individuals (which are not pre-specified).

More info