Teaching activity

Einführung in R, der freie statistische Software (Wintersemester 2008/09)

R is a statistical environment, that allows quick-and-dirty univariate statistical analyses and plots, as well as complex multivariate analyses, bayesian statistics and computational solutions. It has become a de facto standard of statistics.
This course covers the R instructions to produce the most used univariate statistical techniques, as well as principal component analyses. It is good that attendants have followed a basic statistics course, though the several methods and concepts will be reviewed and illustrated during the lessons.

More information per e-mail or here

The statistical analysis of multivariate and compositional data (Sommersemester 2007/08)

Did you notice that your data sets have lots of variables? and that they are typically never negative? Or that they are mostly given in relative units, like ppm, g/l, wt% or mol%? They are compositional data. Compositional data, just due to these characteristics, cannot be analysed naively with standard techniques. This course will present in detail the basic multivariate statistics, how to apply them and interpret their results, with a special focus on compositional data sets.


Aitchison, J. (1986) The Statistical Analysis of Compositional Data. Reedited by: The Blackburn Press, Cadwell, NJ (US). 416 p. ISBN: 1-930665-78-4

Buccianti, A., Mateu-Figueras, G., Pawlowsky-Glahn, V., Eds. (2006) Compositional Data Analysis in the Geosciences; From Theory to Practice. The Geological Society. London (UK). 212 p., ISBN: 1-86239-205-6

(lecture notes of the course of Sommersemmester 2006/07)

Geostatistics: basic theory and practice (Wintersemester 2007/08)

Geostatistics is the name given to a set of statistical techniques to analize spatially-referenced data (i.e. we know where the samples were collected) and interpolate them, assessing the quality of this interpolation.

More information per e-mail or here (attention, only available when I'm in the office)

Multivariate statistics: an introduction with “R” (Wintersemester 2007/08)

This course introduces the most useful statistical methods to analyse data sets with multiple variables. The basics of each method are presented, detailing its usefulness, assumptions, requirements, weaknesses and typical results. The free statistical software R is used to do the analysis.

The statistical analysis of compositional data (Sommersemester 2006/07)

Goal: Introduction to the theory and practice of the statistical treatment of compositions (multidimensional observations of positive components, where each part tells us the importance of a part in a whole).

Lecture Notes on Compositional Data Analysis