Studies and Teaching
In close cooperation with the Chair of Statistics and Data Analytics, the Teaching Unit for Computational Statistics and Mathematics offers a wide range of courses in the fields of statistics and econometrics. In addition to basic methods of descriptive and inferential statistics, the main focus is on multivariate estimation and testing – especially in the context of regression analysis.
The core competence relevant for research and practice lies in the fact that methodological understanding is closely interlinked with the corresponding skills for the computer-aided implementation of statistical methods. Students are enabled to implement and interpret statistical procedures empirically and to evaluate corresponding analyses.
Important notes
- There are no certificate exams (Scheinklausuren) offered in the field of statistics.
- Students of Cultural Studies should contact the Chair of Methods of Empirical Social Research (Prof. Ingo Rohlfing, PhD) if they have any questions regarding classes in statistics or methodology (including the recognition of academic achievements at international faculties).
- The courses "Computational Statistics - Regression in R" and "Computational Statistics - Statistical Learning in R" are open to students from other Master's programmes. After successfully passing the exam, they will receive a certificate of their acquired knowledge.
- Mathematik für Wirtschaftswissenschaftler (Mathematics for economists, module no. 35400, held in German) [5 ECTS]: Students learn the basic mathematical skills required for business studies. By actively solving exercises and practical examples independently, you will learn how to transfer the techniques presented in the lecture to economic problems.
- Statistik 1 und 2 für Wirtschaftswissenschaftler (Statistics 1 and 2 for economists, modules no. 35600a and 35600b, held in German) [10 ECTS]: Descriptive statistics and exploration of data; basics of probability calculation; random variables; discrete and continuous distributions; random samples; point and interval estimates; distribution-bound and distribution-free hypothesis tests; linear regression analysis; the use of standard statistical software.
- Einführung in die Ökonometrie (Introduction to Econometrics, module no. 35555, held in German) [5 ECTS]: The central topic of the course is regression analysis, with which data-based economic relationships can be quantified and corresponding hypotheses tested. The degree of uncertainty underlying the results can be estimated.
- Einführung in die Zeitreihenanalyse (Introduction to time series analysis, module no. 35560, held in German) [5 ECTS]: The module is designed as a basic event on the classical topics of time series analysis - such as level, trend, seasonal and cycle analysis. The first part of the module deals with intuitive, semi- and non-parametric methods, including the simple component model and various smoothing methods. The second part of the course introduces the theory, selection, estimation and diagnostics of ARIMA models. These still play a central role in the application of time series models in practice.
- Computergestützte Statistik - Einführung in R (Computational Statistics - Introduction to R, module no. 35620, held in German) [3 ECTS]: The central topic is the introduction to the work with the statistics program R. In addition to teaching basic programming techniques (objects, functions, loops, etc..), this also includes an introduction to statistical data analysis (creating helpful tables and graphs, descriptive analyses, model estimates).
- Econometric Methods, module no. 35777 [5 ECTS]: The course is the basis of the Master's training in the field of cross-sectional data. Among others, the following topics are covered: deeper interpretations of the least squares (KQ) method and its statistical properties; exact versus asymptotic methods; generalized KQ procedures (GLS, FGLS, 2SLS, IV and others); model validation and specification procedures.
- Paneldatenanalyse (Panel data analysis, module no. 35610, held in German) [5 ECTS]: The central object is the estimation of regression models for panel data. In addition to basic estimation methods, fixed-effects and random-effects estimation are also dealt with. Furthermore, test and prediction methods for the panel data context are treated.
- Seminar Applied Statistics [7 ECTS] (module no. 35802): The computer-based application of statistical methods and the interpretation of the empirical results obtained are core competencies in various professional fields. The aim of this seminar is to acquire these competences with changing thematic focuses that can be assigned to the fields of time series forecasting, microeconometrics (e.g. marketing and capital market research) and robust methods.
- Multivariate Verfahren (Multivariate methods, module no. 35500, held in German) [5 ECTS]: Multivariate methods are an important part of empirical research practice, e.g. in the field of market research. This module deals with basic analysis techniques for multivariate data structures and their theoretical foundation. In addition to an introduction to the basics of multivariate analysis methods, the following topics are covered: Main component analysis, regression analysis, factor analysis, variance analysis, discriminant analysis, cluster analysis.
- Computational Statistics - Regression in R (module no. 35621) [3 ECTS]: The central subject is the estimation of regression models as well as model diagnosis/validation. In addition to graphical methods and classical validation methods and tests, simulation-based approaches are also discussed. The modelling of different scale levels and variable transformations will be discussed. In addition to cross-sectional data, time series and panel data are also referred to.
- Computational Statistics - Statistical Learning in R (module no. 35622) [3 ECTS]: Statistical Learning sums up methods from computational statistics that are designed to deal with high dimensional, complex data sets. Various topics that facilitate modeling and gaining a deeper insight into high dimensional, complex data sets are introduced. Basic linear and nonlinear classification and regression techniques (e.g., lasso, trees, random forests, boosting, support vector machines) and their underlying principles are presented, applied, and discussed in class. Meta-parameter selection, model evaluation, and specification choice in practical settings are also covered in the course.