The webinars will take place on Zoom and a link to the virtual room will be sent out to all those who registered at the registration page.
Speaker: Rayan Ayari (Zeppelin University)
Title: Beyond the Efficient Frontier and 1/N: How to Beat the Market with Deep Reinforcement Learning from Daily Updated Information
Abstract: This paper presents a new approach to the risk-return optimization in continuous portfolio construction using Deep Reinforcement Learning by including higher-order moments of return distributions and enhancing applicability with locally integrable loss functions. We model stock returns by a Markov Decision Process, employing the Twin Delayed Deep Deterministic Policy Gradient algorithm within a specialized trading strategy that accommodates many assets and mimics real-world financial decisions. We backtest S\&P100 constituents using daily data from 2015-2021 and show our model's superiority. Furthermore, we assess the impact of extended features on portfolio performance and analyze allocation behaviors heuristically.
This series of webinars addresses all interested people in probability, stochastic analysis, control, risk evaluation, statistics, with a view towards applications, in particular to renewable energy markets and production. This series brings together the major research themes of the projects STORM, SCROLLER, and SPATUS.