Introduction to data science methods and coding, and their applications to finance
Course Prerequisites
Basic knowledge of statistics, basics knowledge of computer programming
Teaching Methods
The teaching methods consist of three types of lectures: 1) theoretical lectures by means fo slides commented by the instructor; 2) practical lectures by means of software code commented and executed by the instructor; 3) supervised self learning, in which students execute the software code with the supervision of the instructor. The teaching materials consist of selected chapters from the reference book: James, Hastie, Tibshirani: An introduction to statistical learning, with applications in Python, 2023. The chapters from the book will be supplemented with software codes and data, from the same reference book, necessary to reproduce the content of the chapters.
Assessment Methods
The final exam consists in a report elaboration on a database that will be distributed in class. In the report, students should focus on the data science methods introduced in the course, and prove to have a full comprehension of statistics for finance; they will also have to integrate the analysis by providing economic and financial interpretation. The final report format is made of a presentation file (power point or pdf) highlighting the used models, the elaborations and the obtained results. The presentation will be supplemented by a discussion on the topics presented in class. Working students have to contact the lecturer directly in order to make all the necessary arrangements, including course study material, data and documentation for the exam.
Texts
James, Hastie, Tibshirani: An introduction to statistical learning, with applications in Python, 2023.
Contents
Linear regression, logistic regression, regularisation methods, tree models, neural networks, deep learning, model comparison, applications to predictive problems and risk measurement for finance and financial technologies (fintech)
