A1723
Title: Deep hedging under CRRA preferences
Authors: Mahoro Fukui - Keio University (Japan) [presenting]
Abstract: Deep reinforcement learning is leveraged to calculate option prices under the Constant Relative Risk Aversion (CRRA) utility function in heterogeneous markets. Two-Stage Optimization (TSO) is proposed to address challenges that may have prevented traditional deep hedging models from implementing hedging strategies under CRRA preferences. Since TSO does not require a performance measure to be cash-invariant or to take an exponential form, it can be generalized to broader, non-CRRA preferences as well.
