Approximation Rates for Deep Calibration of (Rough) Stochastic Volatility Models

TL;DR

This paper establishes that deep neural networks can efficiently approximate option prices in high-dimensional stochastic volatility models, including rough models, without suffering from the curse of dimensionality, with errors arbitrarily small.

Contribution

It provides the first quantitative error bounds for DNN approximation of option prices in general Markovian and rough stochastic volatility models, showing sub-polynomial network size growth.

Findings

DNNs can approximate option prices with arbitrary accuracy.

Approximation error bounds are dimension-independent.

Network size grows sub-polynomially with dimension and accuracy inverse.

Abstract

We derive quantitative error bounds for deep neural networks (DNNs) approximating option prices on a $d$-dimensional risky asset as functions of the underlying model parameters, payoff parameters and initial conditions. We cover a general class of stochastic volatility models of Markovian nature as well as the rough Bergomi model. In particular, under suitable assumptions we show that option prices can be learned by DNNs up to an arbitrary small error $\varepsilon \in (0,1/2)$ while the network size grows only sub-polynomially in the asset vector dimension $d$ and the reciprocal $\varepsilon^{-1}$ of the accuracy. Hence, the approximation does not suffer from the curse of dimensionality. As quantitative approximation results for DNNs applicable in our setting are formulated for functions on compact domains, we first consider the case of the asset price restricted to a compact set, then we extend these results to the general case by using convergence arguments for the option prices.