TECH6 is taught in the fifth semester, which is focused on market and business analytics. It builds on students’ knowledge and skills in mathematics (TECH1 and TECH4), statistics (TECH3), and programming (TECH2 and TECH5). Its main goal is to strengthen core competencies needed in data-driven analyses of economic activity and familiarize students with a battery of methods used for prediction and establishing causal relationships in experimental and observational data. TECH6 is also a bridge to the sixth semester on data-driven decisions for sustainable value creation and a foundation for more advanced courses in econometrics, time series analysis, and machine learning taught at the Master’s level.
The first part of this course familiarizes students with econometric techniques at an intermediate level. It provides a deep understanding of modern econometric methods that are widely applied across many fields of economics, with a strong emphasis on both theoretical foundations and practical empirical work. All empirical exercises and assignments will be conducted using R, giving students hands-on experience with one of the most widely used programming languages in applied economics research.
The course covers core topics in causal inference and program evaluation, including identification, randomization, instrumental variables, difference-in-differences, regression discontinuity, and matching methods. Students will develop an understanding of when and how to apply each method, as well as the assumptions underlying their validity. The course also addresses more specialized applied topics such as bad controls, measurement error, and clustering, equipping students with the tools to critically assess empirical work.
Finally, students will explore real-world applications across areas such as labour markets, education, and urban economics, reinforcing the practical relevance of the methods covered throughout the course.
The second part of this course provides a comprehensive and practical introduction to supervised and unsupervised machine learning. Students will learn how to handle large datasets and automate the retrieval of new data from internet sources using techniques such as web scraping. The curriculum explores foundational models for both regression and classification, equipping students with the theoretical understanding and practical tools needed to evaluate model performance across a variety of metrics.
All practical implementations in this part of the course are conducted in Python, with a strong emphasis on industry-standard libraries including pandas and scikit-learn (sklearn). To prepare students for modern, real-world data science workflows, the course also covers essential environment management and version control tools, focusing on Anaconda, GitHub, and VS Code. Additionally, students will learn how to optimize their models through hyperparameter tuning and cross-validation, and will explore the integration of AI tools within VS Code to maximize their coding effectiveness and efficiency.
Course outline:
Part 1: Econometric Methods and Applications
- Core topics in causal inference and program evaluation: identification, randomization, instrumental variables, difference-in-differences, regression discontinuity, and matching.
- Testing and inference.
- Hands-on empirical exercises using R.
- Real-world applications in labour markets, education, and urban economics.
Part 2: Introduction to supervised and unsupervised machine learning
- Methods for unsupervised and supervised machine learning implemented in scikit-learn
- Regression models (Ridge, Lasso, Elastic Net)
- Classification models (Logistic regression, Support Vector Machines, Random Forest)