Neural Networks in Finance (R-Module)
Exam number: 6855
Semester: From 2nd semester
Duration of the module: One block
Form of the module (i.e. obligatory, elective etc.): Elective
Frequency of module offer: Every winter semester, usually 1st block
Prerequisites: Students must have taken at least one course that introduced data analysis and statistical and/or machine learning, for example “Data Analysis and Visualization with R” or “Machine Learning with R”, among others. Furthermore, students have to bring along good knowledge of the R programming language and deep interest in quantitative methods and neural networks.
Applicability of module for other study programmes:
Obligatory or elective in other study programmes. For further information check regulations of the study programme.
Person responsible for module: Prof. Dr. Sven Husmann
Name of the professor: Ivan Semeniuk, Dr. Antoniya Shivarova, Dr. Rick Steinert
Language of teaching: English
ECTS-Credits (based on the workload): 6
Workload and its composition (self-study, contact time):
Contact time: 28 h, self-study: 152 h
Contact hours (per week in semester): 4
Methods and duration of examination:
A seminar paper on the research topic of neural networks in finance and an accompanying oral presentation.
Emphasis of the grade for the final grade: Please check the regulations of the study programs.
Aim of the module (expected learning outcomes and competencies to be acquired):
Students enrolled in this course gain in-depth knowledge of advanced neural networks classes and their application on selected financial problems of high scientific and practical relevance. At the beginning of the seminar, all students receive an introduction to the general research topic. Afterwards, groups of two or three seminar participants work independently on a selected empirical research project over several weeks. At the end of the seminar, each group summarizes its findings in an academic paper and gives an oral presentation about the achieved results. In the seminar, the participants acquire skills in literature and data research, improve their programming skills in R and are especially enabled to do in-depth empirical research.
Contents of the module:
Fueled by the vast improvements in computational power and increased data availability of the recent years, neural networks belong to the machine learning algorithms that have become increasingly popular both in scientific research and the industry. Due to the quick development and extensive research in the field, different advanced classes of neural networks, such as CNN, RNN, LSTM among many others, have been shown to achieve promising state-of-the-art results in many application areas such as image and speech recognition, natural language processing, and anomaly detection. Because of the success in these fields, interest in the implementation and application of neural networks in the global financial services industry is growing, for example to improve financial operations, such as credit assessment, optimal stock trading systems and portfolio management. In this seminar, students will be introduced to a particular class of neural networks (CNN, RNN, etc.) that may change each semester. Thereupon, students will apply the models on a typical financial problem, for example for predicting stock movements, among others. Please, check the Moodle course for more details.
Teaching and learning methods:
In the first two weeks, compulsory lectures introduce the general topic. Afterwards, project work is conducted in small groups.
Literature (compulsory reading, recommended literature):
Literature will be announced in the Moodle course.
Self-enrollment in the Moodle course is required. Students must register for the seminar respecting the deadline specified in Moodle.