 |
| Detailed information |
| Pre-requisites |
(*)keine
|
| Original study plan |
Bachelor's programme Statistics and Data Science 2025W |
| Learning Outcomes |
Competences |
| Students are able to perform regression analyses for non-Gaussian response variables, particularly binary, ordinal and multinomial logistic regression as well as regression for count response variables. They can interprete the results and assess the fitted model.
|
|
| Skills |
Knowledge |
- Knowledge and understanding of the basic problems, terms and methods for regression analysis of non-Gaussian responses (k1,k2)
- Applying and critically evaluating of regression models for non-Gaussian responses (k3,k4, k5)
- Fitting, model choice and residual analysis of regression models for non-Gaussian responses with the statistics software R (k3)
- Implementing and performing simulation studies for generalized linear models (k2,k3)
|
- Problems of using linear regression analysis for non-normal response variables
- Regression analysis of binary response variables (logit and probit models)
- Poisson regression
- Regression analysis of positive continuous response variables
- Modelling assumptions and statistical inference for generalized linear model
- Regression analysis of ordinal and multinomial response variables
- Latent utility representation of binary, ordinal and multinomial logistic models
- Model choice and residual analysis for generalized linear model
- Practical regression analysis of generalized linear models in R
|
|
| Criteria for evaluation |
Homework and written exam
|
| Methods |
presentation by the lecturer
presentation of the homework by students and discussion
|
| Language |
German |
| Study material |
Fahrmeir L., Kneib T., Lang S. and Marx B., Regression. Models, Methods and Applications. Springer, 2013
|
| Changing subject? |
No |
| Corresponding lecture |
(*)ist gemeinsam mit 551STMEMVVK14: KV Multivariate Verfahren (4 ECTS) äquivalent zu
4MSMV1KV: Multivariate Verfahren I (8 ECTS)
|
|
