Topics in Advanced Econometrics
Goal and Content:
The course Topics in Advanced Econometrics is designed for master students in their final year and PhD students in their first year. Generally, the goal of the course is to give students insights on recent developments (‘hot topics’) in econometrics which are not subject to the standard canon of econometrics courses in the master’s program.
This year’s course on Topics in Advanced Econometrics covers in selective manner problems which are centered around the issues related to the use of model selection, machine learning, big data, and statistical inference in economics and beyond. The goal of the course is to give students a broader view on the pitfalls and opportunities of doing empirical work in economics using modern machine learning techniques and big data and to make students acquainted with possible solutions to numerous unsolved questions.
The course will be organized as a journal club. During the semester course participants have to give three presentations on specific core papers related to the overall topic of the course.
I: Fundamental Issues of Statistics and Machine Learning
- Statistics and Machine Learning: The Two Cultures
- Prediction vs Explanation
- Big data in Economics
II. The Statistical Crises
- The Reproducibility Crises
- False Discoveries
- The p-value Debate
III. Algorithmic Selection and Inference when Data are massive
- Distributional Properties under Algorithmic Model Selection
- High Dimensional Inference
IV. Machine Learning for Causal Studies
- Estimating Causal Effects by Machine Learning
- Treatment Heterogeneity and Personalized Treatments
- Policy Learning
|Friday||11:45 - 13:15||H305||W. Pohlmeier|
The course material will be provided via ILIAS. You have to register via ZEUS and will then be automatically registered for ILIAS.