Informed Trading model with quantum walks

In the world of finance, the movement of a stock price is often treated as a mystery, a fog of uncertainty where no single trader can predict the future with certainty. Yet, some traders possess an edge: they know something about the asset's future value that others do not. This is the realm of informed trading, a concept that economists have long tried to model to understand how private information seeps into public prices. The standard framework for this is known as the Kyle model, a theoretical setup where a savvy insider, a crowd of random noise traders, and a market maker who sets prices all interact. In this classic scenario, the market maker tries to guess the true value of an asset by looking at the total volume of orders, while the insider tries to hide their knowledge within that volume to maximize profit. For decades, these models assumed that the future value of an asset drifted like a random walker, stepping left or right with equal chance, much like a person wandering aimlessly down a street.

But what if the rules of that walk were different? What if the underlying uncertainty of the market followed the strange, counter-intuitive laws of quantum mechanics rather than the predictable laws of classical physics? This is the question a team of researchers from the Centre for Quantum Technologies and the School of Computing at the National University of Singapore set out to answer. They took the established Kyle model and replaced the standard random walk with a continuous-time quantum walk. In a quantum walk, the "walker" does not simply choose a direction; it exists in a superposition of possibilities, exploring many paths simultaneously before settling on a position. By applying this quantum behavior to the future value of an asset, the researchers simulated a market where the very nature of uncertainty is fundamentally different. Their work does not claim that real-world markets are currently quantum, but rather explores how a market would behave if the asset's future value were governed by quantum dynamics, revealing surprising differences in how prices form and how much profit an insider can make.

The researchers began by setting up a digital simulation of this quantum market. They imagined an asset whose future price could be any point on a line, determined by the position of a quantum walker after a certain amount of time. In their model, the insider knows exactly where the walker will end up, while the market maker only sees the total volume of trades, which is a mix of the insider's orders and the random noise of other traders. The goal was to find a stable state, or equilibrium, where the insider's strategy to maximize profit and the market maker's strategy to set a fair price perfectly balance each other. In the classical version of this model, where the asset's value follows a standard random walk, the results were familiar. The insider adopted a simple, straight-line strategy: if the future price is high, they buy a proportional amount; if it is low, they sell. The market maker's pricing rule was similarly smooth, resembling a gentle curve that adjusted prices based on order volume. In this classical world, the insider's profit grew with time, but it did so slowly, lagging behind the passage of time in a sub-linear fashion.

When the researchers switched the simulation to the quantum walk, the results became dramatically more complex and unexpected. The smooth, straight-line strategies of the classical world vanished. Instead, the insider's optimal strategy fractured into distinct clusters. Rather than buying or selling a little bit for every possible future price, the insider's orders grouped together in specific regions, leaving large gaps in between. Correspondingly, the market maker's pricing rule, which had been a smooth curve, developed sharp, sudden jumps. These jumps aligned perfectly with the gaps in the insider's strategy, creating a pricing landscape that looked like a series of steps rather than a ramp. This non-linearity was a direct consequence of the quantum nature of the asset's future value, where the probability of finding the asset at a certain price is not a simple bell curve but a complex pattern of peaks and valleys.

Perhaps the most striking finding concerned the profitability of the insider. In the classical simulation, the insider's profit grew sub-linearly, meaning that as time passed, the rate of profit accumulation slowed down. However, in the quantum simulation, the insider's profit grew linearly with time. This means that for every unit of time that passed, the insider gained a consistent, steady amount of profit, a much more efficient accumulation than in the classical case. The researchers found that the sensitivity of the market price to order volume, a measure known as the Kyle lambda, also behaved differently. In the classical setting, this sensitivity grew slowly, following a curve that flattened out over time. In the quantum setting, it grew much faster, increasing in direct proportion to time. This suggests that in a market governed by quantum dynamics, the price impact of trades is significantly more pronounced, and the insider can extract value at a much faster rate.

The study relied heavily on numerical methods to reach these conclusions. Because the equations describing the quantum equilibrium were too complex to solve with a simple formula, the researchers used a computer algorithm to iteratively adjust the strategies of the insider and the market maker until they found a stable solution. They ran these simulations for time steps ranging from one to twenty-three, ensuring that the results for the quantum model could be directly compared to the classical model over the same period. The simulations showed that while the classical market remained relatively stable and predictable, the quantum market exhibited a high degree of non-linearity, with the insider's strategy forming multiple distinct groups and the market maker's price jumping between levels. The researchers noted that as the time step increased, the quantum equilibrium became increasingly difficult to compute, requiring careful tuning of the simulation parameters to ensure the results converged.

The implications of these findings are theoretical but profound. They demonstrate that the fundamental nature of uncertainty—whether it follows the diffusive spread of a classical random walk or the ballistic, wave-like spread of a quantum walk—drastically alters market dynamics. In the quantum scenario, the insider does not just have an edge; the very structure of the market allows them to exploit that edge more efficiently, leading to linear profit growth and a pricing rule that reacts with sharp discontinuities. The researchers concluded that this work represents a first step in understanding financial insider information within a quantum setting. While real-world markets are not currently governed by quantum mechanics, the study highlights how introducing quantum concepts into economic models can yield results that are impossible in classical frameworks. It opens a new door for exploring how the laws of physics might one day intersect with the laws of economics, suggesting that the future of financial modeling could involve a deeper integration of quantum theory.