Machine Learning in Finance 5 credits

This course introduces you to data-intensive and computational approaches to financial analysis using the Python programming language. The course aims to showcase you the new possibilities non-linear machine learning methods offer for financial analysis, as well as their caveats, hence supporting you in a career in finance. You will learn to apply machine learning methods to complex datasets, complementing traditional econometric approaches with non-linear data-driven tools. You will learn to master the full pipeline of machine learning-based analysis, including data handling, data preprocessing, feature engineering, supervised and unsupervised learning, model calibration and validation, as well as back-testing. In the course, you will use all the key ML models, including the recent foundational models based on generative AI. The built models are interpreted using explainable AI tools. The course also introduces you to the analysis of unstructured data (for example, natural language processing), causal machine learning, and machine learning-based simulations. Applications are drawn from a broad range of finance and business topics. <br> **Connection to Research ** The course is research-linked in several ways. First, it introduces you to computational and empirical methods that are increasingly used in contemporary finance research, including prediction-oriented modelling, reproducible workflows, and the critical evaluation of model performance. You work with a wide range of datasets and analytical tasks that resemble those used in academic studies of finance. Second, the course strengthens your methodological capabilities relevant for thesis work and advanced empirical finance courses by combining programming, data handling, and model evaluation in a structured workflow. Third, the course introduces you to current research examples in finance, including research themes represented at JIBS. <br> **Connection to Practice ** The course is strongly practice-oriented. You work hands-on with realistic financial datasets and complete tasks that resemble analytical work in finance industry. These tasks include preparing financial data, building predictive models, evaluating out-of-sample performance, interpreting model outputs, and communicating results for decision-making. The emphasis on reproducible coding workflows, critical model validation, and structured problem-solving mirrors professional practice in data-driven finance. The course therefore supports your employability in roles that require analytical reasoning, quantitative modelling, and practical data capability. <br> **Connection to Ethics, Responsibility, Sustainability (ERS) ** Ethics, responsibility, and sustainability are embedded in the course through content. You are expected to reflect on data quality, transparency, reproducibility, and the responsible interpretation and use of machine learning models in finance. The course addresses ethical risks such as biased data, overfitting, opaque models, misleading performance claims, automation without sufficient oversight, and the responsible use of generative AI and other digital tools.