Zurück zur Auswahl
54857-01 - Kolloquium: Machine Learning for Economists and Business Analysts (3 KP)
| Semester | Frühjahrsemester 2025 |
| Angebotsmuster | unregelmässig |
| Dozierende | Anthony Strittmatter (anthony.strittmatter@unibas.ch, BeurteilerIn) |
| Inhalt | This course provides an introduction to predictive and causal machine learning based on the software “R”. Predictive machine learning aims at forecasting the value of an outcome of interest, e.g. sales or turnover, based on observing specific patterns of potentially relevant factors (or “predictors”) like price, quality, weather, advertisement campaigns etc. That is, predictive machine learning “learns” from patterns among predictors in (past) data to forecast the value of the outcome in the future. Causal machine learning aims at assessing the causal effect of some intervention or treatment, like offering or not offering a training program to jobseekers, on an outcome of interest, like employment. The assessment of a causal effect requires that groups receiving and not receiving a discount are comparable in background characteristics which also affect their employment (e.g. previous labor market history, education etc.). Causal machine learning can be used to generate such comparable groups in a data-driven way by estimating two separate models for how the characteristics affect the intervention and the outcome. Such approaches also permit detecting subgroups for whom the treatment effect is particularly large as a function of their observed characteristics (effect heterogeneity analysis). This is useful for optimally targeting specific subgroups by the treatment (optimal policy learning). This course first discusses the underlying assumptions, intuition and usefulness of machine learning for forecasting and causal analysis. It then introduces various machine learning algorithms and discusses their application for prediction/forecasting and causal analysis. Using the statistical software “R” and its interface “R Studio”, these methods are applied to various real-world data sets. The course covers the following topics: • Introduction to the concepts and purposes of predictive and causal machine learning • Basics of predictive machine learning: Model tuning (cross-validation) and performance evaluation (out-of-sample testing) • Prediction based on penalized regression (lasso and ridge regression) • Prediction based on tree-based approaches (trees, random forests) • Causal analysis based on penalized regression (lasso and ridge regression) • Causal analysis using tree-based approaches (causal trees and causal forests) • Causal analysis based on double machine learning • Assessing effect heterogeneity across subgroups • Optimal policy learning to maximize treatment effectiveness using tree-based approaches • Application of all methods to real world data using the statistical software “R” and its interface “R Studio” Lecture slides, R code, and data files will be made available to the course participants. |
| Lernziele | • To understand the ideas, goals, and differences of machine learning for prediction and for causal analysis • To understand the intuition, advantages, and disadvantages of alternative methods • To be able to apply predictive and causal machine learning to real world data using the software “R” and its interface “R Studio” |
| Literatur | For predictive machine learning: G. James, D. Witten, T. Hastie, and R. Tibshirani (2021): An Introduction to Statistical Learning with Applications in R, Springer, New York. Freely available at: https://www.statlearning.com/ For causal machine learning: M. Huber (2023): Causal analysis - Impact evaluation and causal machine learning with applications in R, forthcoming in MIT Press, Cambridge. |
| Teilnahmevoraussetzungen | Basic knowledge of statistics and econometrics (probability theory, conditional means, linear regression). |
| Anmeldung zur Lehrveranstaltung | Registration: Please enroll in the Online Services (services.unibas.ch); Eucor-Students and mobility students of other Swiss Universities or the FHNW first have to register at the University of Basel BEFORE the start of the course and receive their login data by post (e-mail address of the University of Basel). Processing time up to a week! Detailed information can be found here: https://www.unibas.ch/de/Studium/Mobilitaet.html After successful registration you can enroll for the course in the Online Services (services.unibas.ch). Applies to everyone: Enrolment = Registration for the course and the exam! |
| Unterrichtssprache | Englisch |
| Einsatz digitaler Medien | kein spezifischer Einsatz |
| HörerInnen willkommen |
| Intervall | Wochentag | Zeit | Raum |
|---|---|---|---|
| täglich | Siehe Einzeltermine | ||
Einzeltermine
| Module |
Modul: Fachlich-methodische Ausbildung (Promotionsfach: Staatswissenschaften) Modul: Fachlich-methodische Weiterbildung (Doktoratsstudium - Wirtschaftswissenschaftliche Fakultät (Studienbeginn vor 01.02.2024)) |
| Prüfung | Leistungsnachweis |
| Hinweise zur Prüfung | Participation, individual and group home assignments |
| An-/Abmeldung zur Prüfung | Anm.: Belegen Lehrveranstaltung; Abm.: stornieren |
| Wiederholungsprüfung | keine Wiederholungsprüfung |
| Skala | Pass / Fail |
| Belegen bei Nichtbestehen | beliebig wiederholbar |
| Zuständige Fakultät | Wirtschaftswissenschaftliche Fakultät / WWZ, studiendekanat-wwz@unibas.ch |
| Anbietende Organisationseinheit | Wirtschaftswissenschaftliche Fakultät / WWZ |