Machine learning in financial time series forecasting 2400-EN3SL321A
I invite disciplined individuals to this seminar who are prepared for systematic work throughout the entire academic year and, above all, ready for the effort required to prepare a high-quality Bachelor's thesis.
The seminar will cover selected topics regarding the modeling of financial time series, with a particular focus on the following areas:
Market Risk Forecasting and Management: Modeling, estimation, and backtesting of market time series for the purpose of determining Value at Risk (VaR) and Expected Shortfall (ES) measures.
Reinforcement Learning Algorithms in Finance: Utilizing reinforcement learning (RL) for time series analysis, automated investment decision-making, portfolio optimization, and the construction of autonomous trading agents (agent-based modeling).
Behavioral Finance in Market Analysis: Investigating the impact of market anomalies, investor sentiment (sentiment analysis), and crowd psychology on time series anomalies, price volatility, and the formation of speculative bubbles.
Seminar Structure:
1st Semester: Each student is required to review and discuss a selected scientific article related to their planned Bachelor's thesis topic, and in the second part of the semester, present the preliminary results of their empirical research.
2nd Semester: The second semester concludes with a pass/grade obtained after the successful preparation of a high-quality Bachelor's thesis.
Course coordinators
Type of course
Learning outcomes
Knowledge:
1. The student knows the rules of writing a Bachelor’s thesis.
2. The student knows methods of analyzing data and methods of research
3. The student knows basics of methodology of science and ways of formulating scientific hypotheses.
4. The student knows basic definitions connected to hard to measure phenomena.
5. The student has basic knowledge regarding quantitative methods used in research on hard to measure phenomena.
Skills:
1. The student is capable of finding data and information necessary for preparing a Bachelor’s thesis. The student is able to prepare a master thesis according to formal rules.
2. The student has the skill of applying knowledge acquired during studies to problems connected to the topic of his Bachelor’s thesis.
Social competences:
1. The student is capable of finding data and analyzing it. The student can work with data. He is able to formulate a research problem and a scientific hypothesis. The student is able to prepare a Bachelor’s thesis.
2. The student is able to complete the acquired knowledge. He is capable of finding literature.
3. The student is capable of working individually.
Assessment criteria
1st Semester: Assessment based on the presentation and discussion of a selected scientific paper from the scope of the planned thesis, and a presentation of the preliminary research results along with the proposed thesis outline in the second half of the semester.
2nd Semester: Passing the semester is conditional upon the submission and approval of a complete, high-quality Bachelor's thesis (including successful verification through the anti-plagiarism system).
Bibliography
Tailored individually to the student's needs and the specific topic of the Bachelor's thesis (with a particular focus on international scientific journals in the fields of Machine Learning and Financial Econometrics).