In the finance industry, options, or more generally financial derivatives, are contracts that give their owner the right, but not the obligation, to buy or sell some underlying security or set of securities at pre-defined conditions, such as time or price. They’re used in hedging strategies to manage financial risk, or to speculate on future market performance. Just in 2018, more than 13 billion option contracts were traded worldwide.
Due to the variety and complexity of existing option contracts, their performance and prices are most of the time analysed using Monte Carlo simulations. During a Monte Carlo simulation one generates many possible realisations of future scenarios, e.g. stock prices, according to a statistical model or historical data. These scenarios can be analysed and used to estimate, for instance, the probability of certain events, or the expected performance of financial assets.
Although Monte Carlo simulations are generic, they require large numbers of samples to achieve a reasonable accuracy – sometimes on the order of millions of samples, which can take hours or even days – leading to extensive computational costs and preventing fast decision making.
For example, financial derivatives based on (multiple) global (crypto-) currencies may significantly increase the computational cost of estimating future cashflows and foreign exchange risks. Quantum computing may allow to analyse this more efficiently.
Quantum computing, because of its unique exponential compute properties that operate in a completely different way from today’s classical computers, may help significantly increase the efficiency of these calculations by achieving a quadratic speed-up over classical Monte Carlo simulations.
A quadratic speed-up implies that instead of millions of classical samples, we only require a few thousand quantum samples to achieve the same results, which may allow us to get near real-time results.
In a recent research paper on Option Pricing Using Quantum Computers that my colleagues at IBM and I published in collaboration with scientists at JPMorgan Chase, we developed a generic approach to map option contracts to quantum circuits. The algorithms use today’s small, noisy quantum computers to conceptually demonstrate a speed-up for pricing and analysis of the performance of a simple option. This speed-up may allow financial institutions to significantly reduce computation times and cost and increase the number of what-if scenarios and sensitivities that can be analysed in a given time once large enough quantum hardware becomes available.
We are at the very beginning of this new quantum computing journey. Although we are making progress on a number of applications that may benefit from quantum computing, today’s use cases, like our option pricing research, are only the tip of the iceberg.
And while it’s too early to make predictions, we do know that working with quantum computers requires new and different skills. From experts in industries like finance, to individual developers and today’s students, more early adopters need to understand how to use these unique devices and develop quantum algorithms that may lead to an advantage, such as better-managing financial risk as described here.
About Stefan Woerner
Dr. Stefan Woerner is the Global Leader for Quantum Finance & Optimization at IBM Research - Zurich. The focus of his research is the development and analysis of quantum algorithms for optimization, simulation, and machine learning as well as their practical applications, particularly in finance and supply chain management.
About IBM Q
IBM Q is the worlds most advanced quantum computing initiative, focused on propelling the science and pioneering commercial applications for quantum advantage. For more information about IBMs quantum computing efforts, please visit www.ibm.com/ibmq
Every day we send out a free e-mail with the most important headlines of the last 24 hours.
Subscribe now