This book provides the first practical guide to the function and implementation of algorithmic
differentiation in finance. Written in a highly accessible way Algorithmic Differentiation
Explained will take readers through all the major applications of AD in the derivatives setting
with a focus on implementation. Algorithmic Differentiation (AD) has been popular in
engineering and computer science in areas such as fluid dynamics and data assimilation for
many years. Over the last decade it has been increasingly (and successfully) applied to
financial risk management where it provides an efficient way to obtain financial instrument
price derivatives with respect to the data inputs. Calculating derivatives exposure across a
portfolio is no simple task. It requires many complex calculations and a large amount of
computer power which in prohibitively expensive and can be time consuming. Algorithmic
differentiation techniques can be very successfully in computing Greeks and sensitivities of a
portfolio with machine precision. Written by a leading practitioner who works and programmes AD
it offers a practical analysis of all the major applications of AD in the derivatives setting
and guides the reader towards implementation. Open source code of the examples is provided with
the book with which readers can experiment and perform their own test scenarios without
writing the related code themselves.
