Ateneo - UPO

Machine Learning applied to Finance: understanding how machine learning methods can be utilized in practical Financial Markets problems.

Notes, hand-outs and academic/practitioner’s papers.

Artificial Intelligence and Machine Learning techniques have made tremendous progress in the last twenty years, in large part due to the increase in availability of data and in computational capabilities of our computers.
In this course:
you will learn the main machine learning algorithm, in particular supervised and unsupervised;
you will learn both classical Machine Learning techniques and “Deep Learning” ones and the pros and cons of the various approaches;
you will learn to use python, and in particular the SKlearn and Keras libraries, to use this algorithms in practice;
you will explore practical applications of this algorithms to data taken from Financial Markets, learning some of the practical difficulties you need to take into account when doing this.

Basic Python programming, Finance Statistics (basic level).

Lectures, notes, and discussion. The course is designed to be highly interactive and with a practical angle more than a theoretical one. As a consequence at the end of every lesson introducing a Machine Learning technique there will be a practical exercise to put the algorithm to work on a specific problem in the Financial Markets.

In the latest years artificial intelligence and Machine Learning have revolutionized our way of understanding technology and IT, with a large impact on all sectors.
In Financial Markets as well this impact has been felt strong and stronger and there are ever growing useful applications in this area, irrespectively of whether one works in a bank, a traditional asset manager or an hedge fund.
Students with physical disabilities, Learning Disabilities or Special Education Needs can request specific services and tools via the Staff Sviluppo e Coordinamento Carriere e Servizi alle Studentesse e agli Studenti, consulting the University webpage: https://www.uniupo.it/en/services/servicesstudents-physical-or-learning-disabilities Students with disabilities, learning disabilities or special education needs, once they have contacted the University Staff, can refer to the tutor in charge of the course to define the examination modalities.

Collaborative team student project: the students will try their hands on solving a problem, provided by the professor (including data), similar, for complexity and themes, to the ones that have been tackled as exercises during the course. Solving this problem will require the usage of Machine Learning Techniques, using the python libraries SKLearn and Keras, applied to Financial Markets data.

Lesson 1: Supervised Learning – Support Vector Machine, Decision Trees and Random Forests.
Supervised Learning Exercise – using Support Vector Machine to Find the Relation between the Price of a Bond and Swap Rates in the Market.
Lesson 2: Unsupervised Learning – Clustering and PCA.
Unsupervised Learning Exercise – use PCA to Identify the Main Movements of a Government Yield Curve.
Lesson 3: Basics of Neural Networks - Modelling, Activation Functions and Regularization Types.
Neural Network Exercise – use a Deep Neural Network to Price a Vanilla European Option.

Optional (if time allows):
Lesson 4: Advanced Neural Networks Structures – Siamese Networks and AutoEncoders.
Advanced Neural Networks Exercise: Using Autoencoders to Model an Equity Index with a Limited Subset of Stocks.

Upon completion of the course, students will have gained:
a good basic knowledge of the main Machine Learning algorithms;
the ability to use in an autonomous way these techniques in python via SKLearn and Keras;
the knowledge of some practical application of these techniques to the complex world of Financial Markets;