Interest Rate Swap Valuation Using OIS Discounting - An Algorithmic Approach

Abstract

Companies have traditionally valued their interest rate swaps and other financial instruments using LIBOR. However, at the height of the 2008 financial crisis it became evident that LIBOR, which was once considered a proxy for the risk-free rate was no longer adequate as the benchmark reference rate for valuing financial instruments. LIBOR-OIS spread which had hovered around 5 basis points prior to the financial crisis skyrocketed to over 400 basis points in October 2008 thus leaving companies susceptible to counterparty credit risks. Consequently, experts have proposed and industry regulators have endorsed a "dual curve" interest rate curve construction methodology for valuing swaps that are collateralized and centrally cleared. Further, while a vast quantity of literature exists on the topic of OIS discounting, very few, if any, are dedicated to explaining how to implement the new methodology in a practical and reproducible manner. This essay thus seeks to discuss the algorithmic implementation of OIS discounting by drawing heavily from existing literature and by using the Python programming language. The author hopes that this approach will make the topic more accessible to practitioners and students alike and form the basis for further extending the new method to understanding and solving new risk management and quantitative finance challenges.

Keywords: Swaps, Derivatives, Valuation, Yield Curve, OIS, Multi Curve, Dual Curve, Fixed Income, Bonds, Term Structure of Interest Rate