Home Research Teaching Talks About Me

Past Teaching

  • 2024

    • Resampling Techniques and Their Applications (M.Sc. in Bioinformatics, Freie Universität Berlin and M.Sc. in Statistics, Humboldt-Universität zu Berlin)
    • Biostatistics I (M.Sc. in Epidemiology, Institute of Public Health, Charité)

  • 2023

    • Statistics for Small Sample Sizes (M.Sc. in Bioinformatics, Freie Universität Berlin and M.Sc. in Statistics, Humboldt-Universität zu Berlin)
    • Biostatistics II (B.Sc. in Bioinformatics, Freie Universität Berlin)
    • Biostatistics I (M.Sc. in Epidemiology, Institute of Public Health, Charité)

  • 2022

    • Resampling Techniques and Their Application (M.Sc. in Bioinformatics, Freie Universität Berlin and M.Sc. in Statistics, Humboldt-Universität zu Berlin)
    • Biostatistics I (M.Sc. in Epidemiology, Institute of Public Health, Charité)

  • 2021

    • Biostatistics I (M.Sc. in Epidemiology, Institute of Public Health, Charité)
    • Statistics I (B.Sc. Economics, Economics Faculty, Humboldt-Universität zu Berlin)

  • 2020

    • Statistics I (B.Sc. Economics, Economics Faculty, Humboldt-Universität zu Berlin)
    • Statistics II (B.Sc. Economics, Economics Faculty, Humboldt-Universität zu Berlin)

  • 2019

    • Statistics I (B.Sc. Economics, Economics Faculty, Humboldt-Universität zu Berlin)
    • Statistics II (B.Sc. Economics, Economics Faculty, Humboldt-Universität zu Berlin)

Resampling Techniques and Their Applications

M.Sc. Bioinformatics, M.Sc. Data Science (Freie Universität Berlin); M.Sc. Statistics (Humboldt-Universität zu Berlin)
Resampling methods are known as quite robust and precise procedures. We investigate different methods in detail, e.g. drawing with replacements, permutations, wild bootstrap, etc. and answer questions like "When do bootstrap methods work?" and "How do bootstrap methods work?". Besides, we study efficient implementations in the R programming language. In this course, t-test statistics, linear models, generalized linear models and selected rank methods will be resampled using various strategies. The target audience of this course are students wo are interested in modern statistical tools, developed for situations where sample sizes are rather small.

Statistics for Small Sample Sizes

M.Sc. Bioinformatics, M.Sc. Data Science (Freie Universität Berlin); M.Sc. Statistics (Humboldt-Universität zu Berlin)
In this course, we present statistical inference methods to analyse studies with small sample sizes. We study the implications of the assumption "N is large" and try to find an answer for the question "What does large mean?". Inferential methods include the estimation of treatment effects, confidence intervals and hypothesis tests in parametric and nonparametric models. Rank-tests, bootstrap- and permutation procedures are studied in detail. This course aims to increase knowledge about statistical tools for small sample sizes.

Biostatistics I

M.Sc. Epidemiology (Institute of Public Health, Charité - Universitätsmedizin Berlin)
This course introduces students to biostatistical concepts and statistical methods in epidemiological and clinical research. It will enable students to present data descriptively in an appropriate way, understand the basics of statistical tests and interpret their results. Furthermore, students are taught to select appropriate statistical methods for data analysis, calculate and interpret confidence intervals. Finally, students will learn to apply analysis methods such as regression models and survival time models and interpret their results.

Biostatistics II

B.Sc. Bioinformatics (Freie Universität Berlin)

This course builds upon and deepens the fundamental methods introduced in the course "Statistics for Biologists I". It places a strong emphasis on the application of the discussed statistical procedures and methods. Following a brief introductory session that reviews the basics of descriptive and inferential statistics from "Statistics for Biologists I", the following topics will be covered in more detail:

  • PCA/Cluster/Classification
  • Factor Analysis
  • Missing Values
  • Introduction: Non-parametric Statistics, Outliers
  • Multiple Testing/Posthoc Tests
  • Regression - Diagnostic Studies with a Focus on Logistic Regression & ROC
  • Regression - Survival Analysis
  • Regression - Variable Selection
  • Regression - Analysis of Variance (with Repeated Measures)
  • Regression - Mixed Models

For each topic, weekly practice exercises will be assigned, many with a connection to the life sciences. The exercises will be a mix of handwritten and R statistical software (https://www.r-project.org) work. A basic understanding of R programming is required.

Statistics I

B.Sc. Business Administration, B.Sc. Economics (Humboldt-Universität zu Berlin)
This course builds the statistical foundation for students in the economic sciences. It covers descriptive statistics, basic probability theory and random variables.

Statistics II

B.Sc. Business Administration, B.Sc. Economics (Humboldt-Universität zu Berlin)
This course continues the learning of "Statistics I". It covers distributions, statistical tests, regression analysis and basics of time series analysis.