Skip to main content
Top

2020 | Book

The Fitted Finite Volume and Power Penalty Methods for Option Pricing

insite
SEARCH

About this book

This book contains mostly the author’s up-to-date research results in the area. Option pricing has attracted much attention in the past decade from applied mathematicians, statisticians, practitioners and educators. Many partial differential equation-based theoretical models have been developed for valuing various options. These models do not have any practical use unless their solutions can be found. However, most of these models are far too complex to solve analytically and numerical approximations have to be sought in practice.

The contents of the book consist of three parts: (i) basic theory of stochastic control and formulation of various option pricing models, (ii) design of finite volume, finite difference and penalty-based algorithms for solving the models and (iii) stability and convergence analysis of the algorithms. It also contains extensive numerical experiments demonstrating how these algorithms perform for practical problems. The theoretical and numerical results demonstrate these algorithms provide efficient, accurate and easy-to-implement numerical tools for financial engineers to price options.

This book is appealing to researchers in financial engineering, optimal control and operations research. Financial engineers and practitioners will also find the book helpful in practice.

Table of Contents

Frontmatter
Chapter 1. European Options on One Asset
Abstract
In this chapter we first give a brief account of stochastic differential equations governing risky asset/stock dynamics and Itô lemma to be used for deducing the mathematical model of pricing European options on one asset. We then derive the Black–Scholes (BS) equation using the ideal of \(\varDelta \)-hedging and It\(\hat{\mathrm {o}}\)’s lemma. The BS equation is formulated as a variational problem, which is shown to be uniquely solvable. A fitted Finite Volume Method (FVM) is proposed for the discretization of the equation. We prove that the FVM is unconditionally stable and its solution converges to that of the BS equation. Numerical results are presented to demonstrate the usefulness and accuracy of this FVM.
Song Wang
Chapter 2. American Options on One Asset
Abstract
In this chapter, we first give a brief account of the derivation of the differential Linear Complementarity Problem (LCP) governing American put option valuation. We then write this LCP as a variational inequality, which is shown to be uniquely solvable. A partial differential equation with a nonlinear penalty term is proposed to approximate the LCP. We prove that the penalty equation is uniquely solvable and its solution converges to the weak solution to the LCP exponentially. The finite volume method in Chap. 1 is used for the penalty equation. A Newton’s algorithm is proposed to solve the discretized nonlinear system. Numerical experimental results are presented to demonstrate this method produces financially meaningful numerical solutions to the American put option pricing problem.
Song Wang
Chapter 3. Options on One Asset with Stochastic Volatility
Abstract
In this chapter we develop numerical methods for pricing European and American options whose underlying asset price and volatility follow two separate geometric Brownian motions. These methods include a fitted Finite Volume Method (FVM) for the discretization of the resulting 2D Black–Scholes equation and a power penalty approach to the differential Linear Complementarity Problem involving the 2D differential operator of Black–Scholes type. A mathematical analysis will be presented for the convergence of the FVM and power penalty approach. These methods can also be used for pricing options on two assets such as a basket option.
Song Wang
Chapter 4. Options on One Asset Revisited
Abstract
In this chapter we propose a superconvergent Finite Volume Method (FVM) based on that in Sect. 1.​3 for the nonlinear penalized Black–Scholes equation governing the valuation of European and American options on one asset. Unlike the FVM in Sect. 1.​3, we construct an un-symmetric dual mesh using a set of judiciously chosen points. We show that the approximate flux at these points has a 2nd-order truncation error, instead of the 1st-order one in the FVM in Chap. 1. Thus, the resulting FVM has a higher order accuracy at these points, which are called superconvergent points. Numerical results are presented to demonstrate our theoretical findings.
Song Wang
Metadata
Title
The Fitted Finite Volume and Power Penalty Methods for Option Pricing
Author
Prof. Song Wang
Copyright Year
2020
Publisher
Springer Singapore
Electronic ISBN
978-981-15-9558-5
Print ISBN
978-981-15-9557-8
DOI
https://doi.org/10.1007/978-981-15-9558-5

Premium Partner