Machine learning (ML) is changing virtually every aspect of our lives. Today ML algorithms accomplish tasks that until recently only expert humans could perform. As it relates to finance, this is the most exciting time to adopt a disruptive technology that will transform how everyone invests for generations.

In this course, we discuss scientifically sound ML tools that have been successfully applied to the management of large pools of funds. ML in finance is the utilisation of a variety of techniques to intelligently handle large and complex volumes of information, something the finance industry has in excess of. The industry has found many useful applications. Here we will focus on portfolio management and algorithmic trading.

Portfolio management services use algorithms and statistics to automatically establish and manage the investment portfolio of a client. Algorithmic trading is the use of algorithms to conduct trades autonomously.

Olaf Bochmann Phone: +49 159 03784736 Email: bochmann@htw-berlin.de

Olaf Bochmann is an external lecturer at HTW. He has conducted research in Oxford and Cambridge on financial stability and risk management. Recently, he works on data analytics and machine learning problems for the industry. Dr. Bochmann holds a PhD from University Leuven.

Machine Learning In Finance M11/1 WS 2020/21

Goal

Participants will develop a robo-advisor or trade-bot. They will use real financial data to train state of the art models. Further, they will simulate transactions on new data and benchmark a number of performance criteria.

Topics 1. Policy Gradients, Q-learning, Evolutionary Strategy

2. Transformer Models, LSTM, GRU

3. Portfolio Optimization

Resources • GitHub repository for prediction models https://github.com/

OleBo/Prediction-Models-Finance

• Reinforcement Learning: An Introduction (free book by Sutton http://www.incompleteideas.net/book/RLbook2018trimmed.pdf)

• Evolution Strategies as a Scalable Alternative to Reinforcement Learning https://openai.com/blog/evolution-strategies/

Machine Learning in Finance WS 2020/21 M11/1 • Algorithmic Trading • Portfolio Management • Data analytics

Language • English/German

Dates • Introduction Lecture: 7.Oct.

2019 14:00 - 15:30, HTW Berlin Wilhelminenhof, room C442

• Personal tutoring upon request every Monday (14:00 - 15:30)

• 3 more dates for student presentations to be announced. (45 minutes for presentation, followed by 45 minutes discussion and break)

Grading • 40% Projects • 60% Presentation

This is a hands on course. You will need some experience in handling data and in programming. We encourage you to use Python because it provides the best support for libraries. If you are new in Python, take one of the excellent online tutorials in the summer! If you prefer R or Matlab, it’s fine too.

https://medium.com/swlh/5-free-python-courses-for-beginners-to-learn-online-e1ca90687cafhttps://medium.com/swlh/5-free-python-courses-for-beginners-to-learn-online-e1ca90687cafhttps://github.com/OleBo/Prediction-Models-Financehttps://github.com/OleBo/Prediction-Models-Financehttp://www.incompleteideas.net/book/RLbook2018trimmed.pdfhttp://www.incompleteideas.net/book/RLbook2018trimmed.pdfhttps://openai.com/blog/evolution-strategies/