Quantum Computers Now Calculate Complex Financial Options More Efficiently

Florence Paquette and colleagues at the University of Sherbrooke have created a quantum algorithm for pricing discretely monitored lookback options within the Black-Scholes framework. The algorithm reformulates the pricing problem as a quantum evolution process and uses the Variational Quantum Imaginary Time Evolution method to address challenges arising from jump conditions in these path-dependent options. The method offers a key step towards using quantum computers to price complex financial instruments with non-smooth dynamics, potentially providing advantages over classical Monte Carlo simulations.

Quantum algorithms reduce qubit needs for complex option pricing

A sequential quantum formulation, utilising dedicated jump Hamiltonians, reduced qubit requirements by 33% compared to classical Monte Carlo simulations for equivalent accuracy in pricing discretely monitored lookback options. This reduction in qubit count is particularly noteworthy given the limitations of current quantum hardware, where the number of available qubits is a critical constraint. The Black-Scholes model, a cornerstone of modern financial mathematics, typically relies on assumptions of continuous price movements. However, real-world markets often exhibit discrete jumps, sudden, significant price changes caused by events like earnings announcements or geopolitical shocks. Discretely monitored lookback options are path-dependent, meaning their payoff is determined not just by the final asset price, but by the entire trajectory of the underlying asset over a specified period, making them significantly more complex to price than standard European options. The 33% threshold is significant because it suggests the potential for solving complex financial problems on near-term, limited-qubit quantum hardware, something previously unattainable. The application of the Variational Quantum Imaginary Time Evolution method, or VarQITE, successfully prices options with ‘jump’ conditions, sudden changes in value, which are absent in standard option types and pose a strong challenge for traditional modelling. These jumps introduce discontinuities into the pricing partial differential equation (PDE), making it difficult to solve using conventional numerical methods.

Recasting the pricing problem as a quantum state’s evolution bypassed limitations of existing quantum algorithms designed for simpler financial instruments. The core innovation lies in mapping the Black-Scholes PDE onto a Schrödinger-type equation, effectively transforming the financial pricing problem into a quantum mechanical one. This allows the use of quantum algorithms to simulate the time evolution of the option price. Evaluations of the sequential jump Hamiltonian and simultaneous multi-function evolution against classical Monte Carlo simulations revealed that the sequential approach required fewer qubits for equivalent spatial discretisation. Spatial discretisation refers to the process of dividing the range of possible asset prices into a finite number of intervals, which is necessary for numerical computation. The simultaneous method consistently delivered superior accuracy, and circuit complexity, a measure of the number of quantum operations needed, was demonstrably lower with simultaneous evolution for a given level of precision in option pricing. Circuit complexity is a crucial metric as it directly impacts the time and resources required to run the quantum algorithm on a physical quantum computer. Benchmarking indicated that VarQITE, when applied to these complex options, achieved a speedup compared to classical methods considering the resources required to reach a specific accuracy threshold, although the precise magnitude of this speedup varied depending on the option parameters. The VarQITE method approximates the imaginary time evolution operator, which is essential for solving the Schrödinger equation, using a parameterised quantum circuit. This circuit is optimised through a variational principle, minimising a cost function that quantifies the difference between the quantum-computed option price and a reference solution. Despite these promising results, the 33% qubit reduction does not yet account for the significant overhead associated with error correction, a vital step for reliable quantum computation on current hardware. Quantum bits, or qubits, are susceptible to noise and decoherence, which can introduce errors into the computation. Error correction codes are necessary to mitigate these errors, but they require a substantial number of additional qubits.

Quantum computation advances modelling of path-dependent financial derivatives

Pricing these complex financial instruments, contracts whose value hinges on an asset’s entire price history, has long challenged conventional methods. Monte Carlo simulations, a widely used classical approach, can be computationally expensive, especially for high-dimensional problems and path-dependent options. The computational cost of Monte Carlo simulations scales poorly with the number of dimensions and the complexity of the path dependency. A quantum algorithm capable of handling these ‘jump’ conditions represents a significant step towards modelling real-world financial dynamics. However, the current work, while successful in principle, doesn’t yet prove a quantum computer can outperform existing classical techniques; a clear “quantum advantage” remains unproven. Demonstrating a definitive quantum advantage requires showing that the quantum algorithm can solve a problem faster or more efficiently than any known classical algorithm, even with the overhead of error correction.

Even without demonstrating a definitive speed advantage over classical computers, this work is valuable because it tackles a specific and notoriously difficult problem in financial modelling. The team’s success in adapting quantum techniques to handle sudden shifts in price expands the range of financial instruments potentially amenable to quantum solutions. The ability to accurately price options with jump diffusion, a stochastic process that incorporates both continuous price movements and discrete jumps, is crucial for risk management and derivative pricing in volatile markets. A viable quantum approach to pricing discretely monitored lookback options, financial contracts whose value depends on an asset’s price history and incorporates ‘jump’ conditions, is achieved. This opens the door to exploring quantum solutions for other complex financial derivatives, such as Asian options and barrier options, which also exhibit path dependency.

By reformulating the pricing problem as the evolution of a quantum state, existing algorithms designed for simpler financial instruments were bypassed. The traditional approach to option pricing often involves solving the Black-Scholes PDE using finite difference methods or other numerical techniques. These methods can be inefficient for path-dependent options, as they require discretising both time and the path of the underlying asset. Two quantum formulations were proposed, differing in how they model these jumps, and the sequential approach required fewer computational resources. The sequential approach models the jumps as discrete events occurring at specific times, while the simultaneous approach treats the jumps as continuous stochastic processes. This choice of modelling approach impacts the complexity of the quantum circuit and the number of qubits required. This achievement extends the application of Variational Quantum Imaginary Time Evolution, a method for approximating complex calculations on quantum computers, beyond standard option pricing. VarQITE is a hybrid quantum-classical algorithm, meaning it combines quantum computations with classical optimisation techniques. This allows it to leverage the strengths of both quantum and classical computing, making it a promising approach for solving complex problems on near-term quantum hardware.

The researchers successfully developed a quantum algorithm to price discretely monitored lookback options within the Black-Scholes framework. This is significant because accurately pricing options with jump conditions is vital for managing risk and valuing derivatives in fluctuating markets. They achieved this by reformulating the pricing problem as a quantum state evolution and employing the Variational Quantum Imaginary Time Evolution method. The study demonstrates that complex, path-dependent financial options can be handled using a variational quantum framework, potentially extending these techniques to other derivatives like Asian and barrier options.