Speaker: Luca Galimberti (King's College London)
Title: Deep Learning methods for Limit Order Books: an Infinite-Dimensional Perspective
Abstract: Moving from the optimal liquidation problems proposed by Abi Jaber and Neuman (2024), we study operator learning in the context of linear propagator models with transient price impact à la Bouchaud et al. (2004) and Gatheral (2010). Transient price impact persists and decays over time according to some square integrable propagator kernel H.
More specifically, we show that the optimal execution strategy u = u(H) depends continuously on H, therefore giving rise to a highly non-linear continuous solution operator S between suitable infinite-dimensional Banach spaces. In view of an appropriate infinite-dimensional Universal Approximation Theorem, this operator S can in turn be approximated via ad hoc neural architectures, offering a quick to evaluate surrogate of the operator S in rapidly varying market conditions.
The seminar is in the interest of SURE-AI - Sustainable, Risk-Averse and Ethical AI.
This series of webinars addresses all interested people in probability, stochastic analysis, control, risk evaluation, statistics, with a view towards applications, in particular to finance, economy, energy markets and production, but also to physics and biology. The seminar is held in hybrid format, to register to the mailing list of the STAR seminars and receive the link, please write an email to one of the organisers.
This series brings together the major research themes of the group. STAR stands for STochastics And Risk! Welcome!