Deep neural network framework based on backward stochastic differential equations for pricing and hedging American options in high dimensions

  title={Deep neural network framework based on backward stochastic differential equations for pricing and hedging American options in high dimensions},
  author={Yangang Chen and Justin W. L. Wan},
  journal={Quantitative Finance},
  pages={45 - 67}
We propose a deep neural network framework for computing prices and deltas of American options in high dimensions. The architecture of the framework is a sequence of neural networks, where each network learns the difference of the price functions between adjacent timesteps. We introduce the least squares residual of the associated backward stochastic differential equation as the loss function. Our proposed framework yields prices and deltas for the entire spacetime, not only at a given point (e… 

Deep Learning algorithms for solving high dimensional nonlinear Backward Stochastic Differential Equations

The proposed scheme for solving high dimensional nonlinear backward stochastic differential equations (BSDEs) can be stuck in local minima, especially for a complex solution structure and longer terminal time, so it is investigated to reformulate the problem by including local losses and exploit the Long Short Term Memory networks which are a type of recurrent neural networks (RNN).

Neural Options Pricing

This research investigates pricing financial options based on the traditional martingale theory of arbitrage pricing applied to neural SDEs and proposes a variation of the SDE-GAN approach by implementing the Wasserstein distance metric as a loss function for training.

On a Neural Network to Extract Implied Information from American Options

It is shown that machine learning can be used as an efficient numerical technique to extract implied information from American options, particularly when considering multiple early-exercise regions due to negative interest rates.

Quantum-Inspired Tensor Neural Networks for Option Pricing

Tensor Neural Networks (TNN) demonstrate that they can provide significant parameter savings while attaining the same accuracy as compared to the classical Dense Neural Network (DNN).

The One Step Malliavin scheme: new discretization of BSDEs implemented with deep learning regressions

A novel discretization is presented for forward-backward stochastic differential equations (FBSDE) with differentiable coefficients, simultaneously solving the BSDE and its Malliavin sensitivity

An Implementation of Parallel Processing in Pricing and Hedging High-Dimensional American Options

We present a parallel implementation of an existing method for computing prices and deltas of high dimensional American Options as a method of overcoming some of the known issues of computational

Robust deep hedging

We study pricing and hedging under parameter uncertainty for a class of Markov processes which we call generalized affine processes and which includes the Black–Scholes model as well as the constant

Solving optimal stopping problems with Deep Q-Learning

A reinforcement learning approach to model optimal exercise strategies for option-type products by approximate the Q-function with a deep neural network, which does not require the specification of basis functions as in the least-squares Monte Carlo framework and is scalable to higher dimensions.

Learning the random variables in Monte Carlo simulations with stochastic gradient descent: Machine learning for parametric PDEs and financial derivative pricing

A new numerical approximation strategy for parametric approximation problems including the parametric financial pricing problems described above is introduced and it is illustrated by means of several numerical experiments that the introduced approximation strategy achieves a very high accuracy for a variety of high-dimensional parametric approximate problems, even in the L∞-norm.

Pricing and Hedging American-Style Options with Deep Learning

A deep learning method for pricing and hedging American-style options that first computes a candidate optimal stopping policy, then calculates an upper bound, a point estimate and confidence intervals, and constructs an approximate dynamic hedging strategy.



Deep Learning-Based Numerical Methods for High-Dimensional Parabolic Partial Differential Equations and Backward Stochastic Differential Equations

We study a new algorithm for solving parabolic partial differential equations (PDEs) and backward stochastic differential equations (BSDEs) in high dimension, which is based on an analogy between the

Machine Learning Approximation Algorithms for High-Dimensional Fully Nonlinear Partial Differential Equations and Second-order Backward Stochastic Differential Equations

This work proposes a new method for solving high-dimensional fully nonlinear second-order PDEs and shows the efficiency and the accuracy of the method in the cases of a 100-dimensional Black–Scholes–Barenblatt equation, a100-dimensional Hamilton–Jacobi–Bellman equation, and a nonlinear expectation of a 200-dimensional G-Brownian motion.

Asymptotic Expansion as Prior Knowledge in Deep Learning Method for High dimensional BSDEs

It is demonstrated that the use of asymptotic expansion as prior knowledge in the “deep BSDE solver”, which is a deep learning method for high dimensional BSDEs proposed by Weinan et al. drastically reduces the loss function and accelerates the speed of convergence.

Some machine learning schemes for high-dimensional nonlinear PDEs

The proposed new machine learning schemes for solving high dimensional nonlinear partial differential equations (PDEs) rely on the classical backward stochastic differential equation (BSDE) representation of PDEs and provide error estimates in terms of the universal approximation of neural networks.

DGM: A deep learning algorithm for solving partial differential equations

Efficient Hierarchical Approximation of High-Dimensional Option Pricing Problems

The objective of this article is to show how an efficient discretization can be achieved by hierarchical approximation as well as asymptotic expansions of the underlying continuous problem.

On the Use of Policy Iteration as an Easy Way of Pricing American Options

It is demonstrated that policy iteration, introduced in the context of HJB equations in [P. Forsyth and G. Labahn], is an extremely simple generic algorithm for solving linear complementarity problems (LCPs) resulting from the finite difference and finite element approximation of American options.

On the Rate of Convergence of Discrete‐Time Contingent Claims

This paper characterizes the rate of convergence of discrete‐time multinomial option prices. We show that the rate of convergence depends on the smoothness of option payoff functions, and is much

Optimal stopping of Markov processes: Hilbert space theory, approximation algorithms, and an application to pricing high-dimensional financial derivatives

The authors propose a stochastic approximation algorithm that tunes weights of a linear combination of basis functions in order to approximate a value function and prove that this algorithm converges and that the limit of convergence has some desirable properties.

A Sotchastic Mesh Method for Pricing High-Dimensional American Options

A new stochastic mesh method is presented for pricing high-dimensional American options when there is a finite, but possibly large, number of exercise dates and the algorithm provides point estimates and confidence intervals and it converges to the correct values as the computational effort increases.