R is a high-level language and an environment for data analysis and data visualization. While many important statistical methods are already included in the base R installation, the main benefit is its open-source philosophy which makes R highly extensible and renders possible the availability of new, cutting edge applications in many different fields. The popularity of R increased constantly during the last years and by now, it is arguably the most popular software for data analysis in the statistical community.

The course starts with an standard part that focuses on the following:

• An introduction to R
• Dynamic documents with R Markdown and Quarto
• Linear Models: Simple and Multiple Linear Regression, ANOVA/ANCOVA, descriptive statistics and visualization, diagnostics, data transformations, model selection procedures, model plots (effect displays and posterior predictive checks), design matrices/contrasts
• Generalized Linear Models: Binary, Multinomial and Proportional-Odds Logistic Regression, Poisson and Negative-Binomial Regression, odds ratios, maximum likelihood estimation, descriptive statistics and visualization, diagnostics etc. (as specified for the linear models above)

Depending upon students' interests and the data sets they want to analyze, a selection of these additional methods may be covered:

• Mixed-Effects Models
• Decision Trees
• Classification methods: Naive Bayes, k-NN, ...
• Cluster Analysis: Hierarchical, non-hierarchical, parametric/model-based
• Correspondence Analysis
• Principal Components Analysis
• Multidimensional Scaling
• Social Network Analysis
• ...

• Attendance is compulsory. Students have to attend classes for at least 80% of the total time, i.e., 18 of 22.5 hours. If you know you will miss a class, please inform me in advance!

• development of a project concept (10 %)
• written report on the analysis of a dataset using advanced statistical methods (50 %)
• oral presentation of analysis results (40 %)