Introduction to the numerical calculation of financial derivates. Computational finance (Q4954012)

The literature on financial mathematics in general and on pricing of derivatives in particular has grown a lot over recent years. However, none of these books addresses in much detail the issue of numerical computation of the required quantities. The present German textbook was written with this gap in view and the author's goal is to give an elementary introduction to some of the ideas, concepts and techniques behind the numerical computation of option prices. Chapter 1 covers some basics: options, the Black-Scholes PDE, binomial trees as a first numerical way to get single option prices and some basic probabilistic notions related to the Black-Scholes model. The treatment of finance and probability here is kept deliberately short, sometimes even too short. Chapter 2 discusses random number generation, transformations from uniform to general and in particular to normal distributions, and integration with Monte Carlo methods and low discrepancy sequences. Chapter 3 then presents the basic ideas for the numerical solution of stochastic differential equations, following the book by \textit{P. E. Kloeden} and \textit{E. Platen} [ Numerical solution of stochastic differential equations. Berlin, Springer (1992; Zbl 0752.60043)]. Throughout this first half of the book, the emphasis is more on how things work than on why. The book does convey the basic ideas, but often in the form of algorithms or recipes, and refers to other literature for more detailed explanations. In contrast to the primarily stochastic methods in the first half, the book's second half concentrates on the numerical solution of analytically formulated problems from finance. Chapter 4 explains finite difference techniques in the context of pricing European and American options in the Black-Scholes model. Starting from the basic ideas, this includes explicit and implicit methods, a discussion of stability, the Crank-Nicolson scheme and a detailed treatment of boundary conditions. American option pricing is formulated as a free boundary problem, related to obstacle problems and rewritten as a linear complementarity problem. The discretized version of this is then solved with the projection SOR method of Cryer. Chapter 5 gives an introduction to finite element methods. It begins with a general formulation via weighted residuals, explains the ideas by the Galerkin method for hat functions and applies this to American options, reformulated as a variational inequality problem. Basics of error estimates and solution concepts are covered as well. This second half is written very competently with a good balance between theory and applications. A series of six short appendices provides some additional elementary background material. The overall approach of this book (as mentioned in the preface) is one of learning by calculating. In my opinion, this works here very well for the numerical aspects, but the financial background remains a bit too rudimentary. It is advisable to read an introduction to financial mathematics beforehand or at least in parallel in order to maximize the benefits from this book; but then they can indeed be quite substantial.