FINTECH: TECHNOLOGY FOR FINANCE AND INSURANCE

Modern finance cannot exist without the support of technologies originating from other fields. Among these technologies, an increasingly important role is played by Artificial Intelligence, in its various forms. This course aims to provide knowledge on intelligent methodologies and machine learning techniques, generally inspired by the problem-solving abilities typical of higher living beings, for solving problems of interest in the financial and insurance domains. For example: intelligent metaheuristics inspired by the principles of natural evolution and swarm intelligence are presented for solving complex portfolio selection problems; predictive methods inspired by the functioning of the biological brain are introduced; and systems for identifying optimal financial trading policies based on machine learning techniques inspired by the learning modalities of higher living beings are discussed. Additionally, the course introduces and utilizes software tools to implement all the presented methodologies.

1. Knowledge and understanding:
1.1. To grasp the theoretical aspects of intelligent methodologies and machine learning techniques presented in the course;
1.2. To comprehend, apply, and, when necessary, adapt such methodologies and techniques for solving financial problems.

2. Ability to apply knowledge and understanding:
2.1. To identify and apply appropriate intelligent methodologies and machine learning techniques for operational problem-solving;
2.2. To set up problem-solving processes and perform necessary computations using software tools.

3. Judgement skill:
3.1. To interpret financial implications of the computational results;
3.2. To understand the merits and limitations of learned intelligent methodologies and machine learning techniques.

Having clear the contents of the following courses in the undergraduate programs in the economic field: Mathematics-1, Mathematics-2, Computing Skills for Economics. Additionally, having some experience in software programming is advisable.

- Introduction to Artificial Intelligence and to Machine Learninf in finance and in insurance.
- Intelligent metaheuristics for complex optimization and financial/insurance applications.
- Supervised learning (Decision Tree, Random Forest, Perceptron, and Multi Layer Perceptron) and financial/insurance applications.
- Reinforcement Learning and financial/insurance applications.
- Elements of Natural Language Processingand financial/insurance applications.
- Elements of Group Method of Data Handling and financial/insurance applications.
- Implementations in Matlab.

- Teaching materials available at the web page of the e-learning platform Moodle. [Reference materials]

- Alpaydin E. (2014) Introduction to Machine Learning. The MIT Press [Integrative reading]

The exam consists in three homeworks and in an oral examination.
The homeworks: 1) must be carried out in couple; 2) are valid for the whole academic year and not beyond; 3) their carrying out must be sent no later than a pre-established deadline (the way of sending and the deadline will be indicated during the course).
Regarding the oral examination: 1) it must be carried out individually; 2) it is divided into three parts: in the first part one has to critically present a research article; in the second part one has to apply one or more methodologies learned during the course to reply the results presented in the research paper; in the third part, the student must answer a question chosen by the instructor from three proposed by the instructor, related to topics covered in lectures and assigned teaching materials.
Concerning the evaluation: 1) each homework is worth 0 to 3 possible points, for a total from 0 to 12 points; 2) the oral examination is worth 0 to 18 possible points.
The sum of the points obtained from the homeworks and from the oral examination constitutes the final mark.

oral

Evaluation grid:
- 18–22: Incomplete and inaccurate completion of homework assignments; barely sufficient critical thinking skills applied to the understanding of applications, methods, models, and related content presented in the specialized literature; minimal or no ability in the development and implementation of software related to such applications, methods, and models; barely sufficient knowledge of the topics covered in class and in the assigned teaching materials.
- 23–26: Complete but not entirely accurate completion of homework assignments; fair critical thinking skills applied to the understanding of applications, methods, models, and related content presented in the specialized literature; modest ability in the development and implementation of software related to such applications, methods, and models; fair knowledge of the topics covered in class and in the assigned teaching materials.
- 27–30L: Complete and rigorous completion of homework assignments; excellent critical thinking skills applied to the understanding of applications, methods, models, and related content presented in the specialized literature; good ability in the development and implementation of software related to such applications, methods, and models; in-depth knowledge of the topics covered in class and in the assigned teaching materials.

The course is articulated into:
a) lectures;
b) implementation and use of intelligent methdologies through software tools;
c) individual study.
Students are strongly encouraged to actively attend classes.

Site of the course present on the e-learning platform Moodle.

This subject deals with topics related to the macro-area "Human capital, health, education" and contributes to the achievement of one or more goals of U. N. Agenda for Sustainable Development