direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Page Content

There is no English translation for this web page.

Machine Learning Methods for Economists

Dienstag, 24. März 2020

SoSe 2020
Fr. 12:00-14:00, online
Links to Videoconference will be available on ISIS

Registration via QISPOS (details will be announced later)
English language skills, knowledge of and interest in math, statistics and econometrics. Basic knowledge of any programming language is desirable but not required.

The corse is part of the module 70382 Econometrics and Machine Learning (module can be completed in one semester). It focuses on machine learning methods through the paradigm of economic thinking and highlight similarities and differences between classical econometrics and machine learning.

The students get a thorough overview and hands-on experience with modern machine learning methods and their possible economic applications. They obtain solid basic knowledge of current machine learning techniques and are able to apply them to real-life problems. They get experienced with programming in Python and using the state of-the-art machine learning libraries.

Lectures describe theoretical background of different statistical and machine learning techniques: classification and regression trees, LASSO, k-nearest-neighbors, neural networks, support vectors machines etc. as well as general principles: cross validation, out-of-sample accuracy, concepts of probability theory. Tutorials deal with real life economic problems and demonstrate how to apply machine learning techniques to these problems using Python programming language. 

In SoSe2020 the course will combine self-preparation (online lecture slides and papers), homeworks, Q&A sessions and tutorials (in online conference format).

The course is graded based on the final written exam. Successful completing of at least 50% of the homeworks is required to write the exam.

For further information, please contact Prof. Almosova at .




Zusatzinformationen / Extras