Finanza e assicurazioni - Finance and insurance

1. INTRODUCTION TO THE COURSE • Risk management in banking • Financial risks 2. THE ROLE OF ARTIFICIAL INTELLIGENCE IN BANKING AND FINANCE • Machine Learning, Deep Learning, and Data Science • Applications and examples in Banking and Finance (case studies presentation) 3. INTRODUCTION TO PYTHON AND (FINANCIAL) DATA GATHERING • Describing the main Python packages for finance • Describing the main sources of financial data • Analyzing the different database structures • Exploratory data analysis and Descriptive statistics 4. SUPERVISED LEARNING: OVERVIEW • Regression (specific skills for ML applications) • Support Vector Machine • K-Nearest Neighbors • Linear Discriminant Analysis • Classification and Regression Trees 5. MODEL PERFORMANCE • Overfitting and Underfitting • Cross Validation • Evaluation Metrics 6. SUPERVISED LEARNING: REGRESSION • Linear Regression (specific skills for ML applications) • Logistic Regression (specific skills for ML applications) • Polynomial Regression (specific skills for ML applications) • Applications and examples in Banking and Finance (i.e. determinants of bank default; asset price prediction; ...): coding exercises and academic papers discussion 7. SUPERVISED LEARNING: CLASSIFICATION • Classification Predictive Modeling • Binary Classification • Multi-Label Classification • Imbalanced Classification • Applications and examples in Banking and Finance (i.e. loan defaults; fraud detection; ...): coding exercises and academic papers discussion 8. ARTIFICIAL NEURAL NETWORKS • Architecture • Training • Hyperparameters • Applications and examples in Banking and Finance (i.e. determinants of bank default; ...): coding exercises and academic papers discussion 9. UNSUPERVISED LEARNING: OVERVIEW • Dimensionality reduction • Clustering 10. MODEL PERFORMANCE • Evaluation Metrics 11. UNSUPERVISED LEARNING: DIMENSIONALITY REDUCTION • Feature selection • Feature extraction • Applications and examples in Banking and Finance (i.e. risk management; ...): coding exercises and academic papers discussion 12. UNSUPERVISED LEARNING: CLUSTERING • K-means • Hierarchical clustering • Applications and examples in Banking and Finance (i.e. bank business models identi- fication; ...): coding exercises and academic papers discussion 13. NATURAL LANGUAGE PROCESSING • Topic modeling • Sentiment analysis • Applications and examples in Banking and Finance (i.e. trading strategies; crypto sentiment evaluation; ...): coding exercises and academic papers discussion 14. ETHICS AND TRANSPARENCY IN ARTIFICIAL INTELLIGENCE • (How to deal with) applications and examples 15. PROJECT PRESENTATIONS AND CONCLUSION OF THE COURSE • Group presentations • Wrapping up of the course 16. REAL WORLD EXPERIENCES • Lessons from regulators (European Central Bank; Bank of Italy; ...) • Practitioners (Bankers; Financial or industrial companies; ...)

Non ci sono prerequisiti vincolanti. Tuttavia, gli studenti dovrebbero preferibilmente conoscere gli aspetti principali della gestione bancaria (ad esempio la gestione del rischio) e padroneggiare la statica di base per l'analisi dei dati finanziari.

Tatsat, H., Puri, S., & Lookabaugh, B. (2020). Machine Learning and Data Science Blueprints for Finance. O’Reilly Media.

Non obbligatoria

Sistema di valutazione • Esame finale – STUDENTI FREQUENTANTI * Progetto di gruppo (con presentazione in aula) * Esame scritto individuale – STUDENTI NON FREQUENTANTI * Esame scritto individuale • Pesi – STUDENTI FREQUENTANTI * Class participation: 5/30 * Valutazione del progetto di gruppo: 15/30 * Esame scritto individuale: 13/30 – STUDENTI NON FREQUENTANTI * Esame scritto individuale: 33/30

Lezioni frontali e esercitazioni

1. INTRODUCTION TO THE COURSE • Risk management in banking • Financial risks 2. THE ROLE OF ARTIFICIAL INTELLIGENCE IN BANKING AND FINANCE • Machine Learning, Deep Learning, and Data Science • Applications and examples in Banking and Finance (case studies presentation) 3. INTRODUCTION TO PYTHON AND (FINANCIAL) DATA GATHERING • Describing the main Python packages for finance • Describing the main sources of financial data • Analyzing the different database structures • Exploratory data analysis and Descriptive statistics 4. SUPERVISED LEARNING: OVERVIEW • Regression (specific skills for ML applications) • Support Vector Machine • K-Nearest Neighbors • Linear Discriminant Analysis • Classification and Regression Trees 5. MODEL PERFORMANCE • Overfitting and Underfitting • Cross Validation • Evaluation Metrics 6. SUPERVISED LEARNING: REGRESSION • Linear Regression (specific skills for ML applications) • Logistic Regression (specific skills for ML applications) • Polynomial Regression (specific skills for ML applications) • Applications and examples in Banking and Finance (i.e. determinants of bank default; asset price prediction; ...): coding exercises and academic papers discussion 7. SUPERVISED LEARNING: CLASSIFICATION • Classification Predictive Modeling • Binary Classification • Multi-Label Classification • Imbalanced Classification • Applications and examples in Banking and Finance (i.e. loan defaults; fraud detection; ...): coding exercises and academic papers discussion 8. ARTIFICIAL NEURAL NETWORKS • Architecture • Training • Hyperparameters • Applications and examples in Banking and Finance (i.e. determinants of bank default; ...): coding exercises and academic papers discussion 9. UNSUPERVISED LEARNING: OVERVIEW • Dimensionality reduction • Clustering 10. MODEL PERFORMANCE • Evaluation Metrics 11. UNSUPERVISED LEARNING: DIMENSIONALITY REDUCTION • Feature selection • Feature extraction • Applications and examples in Banking and Finance (i.e. risk management; ...): coding exercises and academic papers discussion 12. UNSUPERVISED LEARNING: CLUSTERING • K-means • Hierarchical clustering • Applications and examples in Banking and Finance (i.e. bank business models identi- fication; ...): coding exercises and academic papers discussion 13. NATURAL LANGUAGE PROCESSING • Topic modeling • Sentiment analysis • Applications and examples in Banking and Finance (i.e. trading strategies; crypto sentiment evaluation; ...): coding exercises and academic papers discussion 14. ETHICS AND TRANSPARENCY IN ARTIFICIAL INTELLIGENCE • (How to deal with) applications and examples 15. PROJECT PRESENTATIONS AND CONCLUSION OF THE COURSE • Group presentations • Wrapping up of the course 16. REAL WORLD EXPERIENCES • Lessons from regulators (European Central Bank; Bank of Italy; ...) • Practitioners (Bankers; Financial or industrial companies; ...)

Non ci sono prerequisiti vincolanti. Tuttavia, gli studenti dovrebbero preferibilmente conoscere gli aspetti principali della gestione bancaria (ad esempio la gestione del rischio) e padroneggiare la statica di base per l'analisi dei dati finanziari.

Tatsat, H., Puri, S., & Lookabaugh, B. (2020). Machine Learning and Data Science Blueprints for Finance. O’Reilly Media.

Non obbligatoria

Sistema di valutazione • Esame finale – STUDENTI FREQUENTANTI * Progetto di gruppo (con presentazione in aula) * Esame scritto individuale – STUDENTI NON FREQUENTANTI * Esame scritto individuale • Pesi – STUDENTI FREQUENTANTI * Class participation: 5/30 * Valutazione del progetto di gruppo: 15/30 * Esame scritto individuale: 13/30 – STUDENTI NON FREQUENTANTI * Esame scritto individuale: 33/30

Lezioni frontali e esercitazioni