Neural network approximation for superhedging prices
Abstract
This article examines neural network-based approximations for the superhedging price process of a contingent claim in a discrete time market model. First we prove that the -quantile hedging price converges to the superhedging price at time for tending to , and show that the -quantile hedging price can be approximated by a neural network-based price. This provides a neural network-based approximation for the superhedging price at time and also the superhedging strategy up to maturity. To obtain the superhedging price process for , by using the Doob decomposition it is sufficient to determine the process of consumption. We show that it can be approximated by the essential supremum over a set of neural networks. Finally, we present numerical results.
Cite
@article{arxiv.2107.14113,
title = {Neural network approximation for superhedging prices},
author = {Francesca Biagini and Lukas Gonon and Thomas Reitsam},
journal= {arXiv preprint arXiv:2107.14113},
year = {2021}
}