The seminar is a three-day block seminar at the end of the semester (2.-4.2.2022, 9 am to 4 pm). The seminar is aimed at students who are either in a higher Master's degree year (about to write their Master's thesis) or doctoral students. The seminar is not suitable for Bachelor's students or Master's students at the beginning of their Master's studies, as it already presupposes extensive methodological knowledge. close

In this seminar, methods of advanced quantitative methods for cross-sectional analysis are taught. Topics of the seminar are: A. Properties of statistical estimators B. Regression modelling: - Linear regression (advanced level) - Nonlinear models / generalized linear models - Poisson regression - Logit and probit model - Ordered response models (ordered probit / logit model) - Multinomial response models (multinomial logit model) C. Survey design and survey weights (and how to use them) In this context, we will also talk about model selection, transformation of variables, heteroscedasticity, and endogeneity. We will use the statistical software R (with R-Studio as editor) in the seminar. Either 5 or 10 ECTS points can be achieved in the course. For 5 points, all exercises must be worked on during the seminar and handed in after the seminar. For 10 points, the written elaboration of a research project (max. 12 pages) is necessary, for which the student already brings a topic, a question and an associated data set as a prerequisite. Termin 4.2.2022 vsl. digital close

Suitable literature for the seminar is, e.g. Agresti, Alan. (2017). Statistical methods for the social sciences. Pearson. Mittelhammer, R., & Mittelhammer, R. C. (1996). Mathematical statistics for economics and business (Vol. 78). New York: Springer. Kabaco?, R. I. (2010). R in Action. Manning. Fahrmeir, L., Heumann, C., Künstler, R., Pigeot, I., & Tutz, G. (2016). Statistik: Der Weg zur Datenanalyse. Springer-Verlag. close