Back
| Semester | spring semester 2023 |
| Course frequency | Irregular |
| Lecturers | Martin Huber (mar.huber@unibas.ch, Assessor) |
| Content | 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). Finally, by repeatedly assigning alternative treatments over time in an appropriate way, one may learn and converge to the assignment of the most effective treatment (reinforcement 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, bagging, random forests) • Further predictive machine learners: boosting, support vector machines, neural networks (deep learning), and ensemble methods • 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 • Reinforcement learning to learn the most effective treatment (among several alternatives) by repeated treatment assignment over time • 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 via e-mail. |
| Learning objectives | • 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” |
| Bibliography | 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. A draft version in pdf format will be made available to the course participants via e-mail. |
| Admission requirements | Basic knowledge of statistics and econometrics (probability theory, conditional means, linear regression), basic command of the statistical software “R”. |
| Course application | Registration: Please enrol in MOnA. EUCOR-Students and students of other Swiss Universities have to enrol at the students administration office (studseksupport1@unibas.ch) within the official enrolment period. Enrolment = Registration for the exam! |
| Language of instruction | English |
| Use of digital media | No specific media used |
| Course auditors welcome |
| Interval | Weekday | Time | Room |
|---|---|---|---|
| wöchentlich | Tuesday | 08.30-12.00 | Wirtschaftswissenschaftliche Fakultät, Seminarraum S15 HG.31 |
| Comments | Dozent: Martin Huber (Uni Fribourg) |
| Date | Time | Room |
|---|---|---|
| Tuesday 25.04.2023 | 08.30-12.00 | Wirtschaftswissenschaftliche Fakultät, Seminarraum S15 HG.31 |
| Tuesday 02.05.2023 | 08.30-12.00 | Wirtschaftswissenschaftliche Fakultät, Seminarraum S15 HG.31 |
| Tuesday 09.05.2023 | 08.30-12.00 | Wirtschaftswissenschaftliche Fakultät, Seminarraum S15 HG.31 |
| Tuesday 16.05.2023 | 08.30-12.00 | Wirtschaftswissenschaftliche Fakultät, Seminarraum S15 HG.31 |
| Tuesday 23.05.2023 | 08.30-12.00 | Wirtschaftswissenschaftliche Fakultät, Seminarraum S15 HG.31 |
| Tuesday 30.05.2023 | 08.30-12.00 | Wirtschaftswissenschaftliche Fakultät, Seminarraum S15 HG.31 |
| Modules |
Modul: Fachlich-methodische Weiterbildung (Doctoral Studies - Faculty of Business and Economics) |
| Assessment format | record of achievement |
| Assessment details | Multiple choice exam at the end of the course: 15.06.2023; 09:00-10:00; WWZ S14: A-Z. |
| Assessment registration/deregistration | Reg.: course registration, dereg: cancel course registration |
| Repeat examination | no repeat examination |
| Scale | Pass / Fail |
| Repeated registration | as often as necessary |
| Responsible faculty | Faculty of Business and Economics , studiendekanat-wwz@unibas.ch |
| Offered by | Faculty of Business and Economics |