Data Analytics and Business Economics: Machine Learning from a Regression Perspective

You will learn the fundamentals of machine learning with a focus on methods that extend or build upon regression analysis, tailored to problems in business and economics. You will learn how data-driven models can make accurate predictions, for example identifying potential customers for targeted advertising or matching online users with relevant offers.

You will acquire key skills such as preparing and splitting data, choosing and tuning models, balancing accuracy and interpretability, guarding against overfitting with proper validation, handling many predictors and nonlinear patterns, and communicating results for decision-making.

We take a unified, regression-first view of traditional machine learning algorithms, so we connect new tools to concepts you already know. By the end you will build end-to-end prediction pipelines in Python, compare models fairly, develop knowledge of the formal mechanisms behind machine learning algorithms for tabular data and understand when a simple approach beats a complicated one.

Course literature

The course literature listed may be updated up to eight weeks before the course begins.

Teaching is a blend of video lectures, guided exercise classes and computer labs. The video lectures introduce you to concepts and methods, while weekly guided exercise sessions provide you with the opportunity to deepen understanding through problem-solving and discussion. The computer labs focus on programming and data analysis in Python, allowing you to apply methods directly to real or simulated datasets.

Your learning process combines independent preparation with collaborative in-class work to reinforce key ideas. Assessment is based on a final written exam, completion of the programming assignments, and active participation in the weekly exercise classes. 

The course is designed in a way where you are expected to have basic coding skills in Python, statistical knowledge corresponding to an undergraduate course in mathematical statistics, and familiarity with matrix algebra.

Prerequisites

Students admitted to the master programme Data Analytics and Business Economics are eligible for this course. Students admitted to the Master Programme in Economics with at least 30 ECTS-credits in economics at the advanced level including Advanced Econometrics are eligible for the course. For other students, a Bachelor degree including at least 30 ECTS-credits in statistics of which 7.5 ECTS-credits in econometrics or regression analysis, or a Bachelor degree in economics or business administration with at least 15 ECTS-credits in statistics of which 7.5 ECTS in econometrics or regression analysis is required.