Wiener and Poisson Processes
From Continuous Random Motion to Discrete Events in Financial Markets
Financial markets are often described as complex, adaptive systems shaped by the interaction of millions of participants, evolving information sets, and constantly shifting expectations. Despite this complexity, much of quantitative finance is built upon a surprisingly small set of mathematical tools designed to model uncertainty and randomness. Among the most fundamental of these tools are the Wiener process and the Poisson process.
These two stochastic processes represent distinct ways of thinking about how events unfold through time. The Wiener process models continuous uncertainty, where change occurs through an accumulation of countless small fluctuations. The Poisson process models discrete uncertainty, where events occur suddenly and individually at random points in time. Together, they form the foundation of much of modern probability theory, financial mathematics, risk modelling, and quantitative research. Understanding their differences, and their limitations, provides valuable insight into how markets behave and how uncertainty can be represented mathematically.
The Challenge of Modelling Randomness
Financial markets are driven by uncertainty. Future prices depend on information that has not yet arrived, decisions that have not yet been made, and interactions between participants whose behaviour cannot be perfectly predicted.
To model such systems, mathematicians require processes capable of describing how uncertainty evolves through time. A useful stochastic process must capture not only where a system currently exists but also how it may evolve into the future; the Wiener and Poisson processes achieve this in fundamentally different ways. The Wiener process describes a world of continuous random motion; whereas, the Poisson process describes a world of random events.
Both appear throughout finance, economics, engineering, physics, and many other disciplines because they capture two of the most important forms of uncertainty observed in nature and society.
The Wiener Process
The Wiener process, often referred to as Brownian motion, is named after mathematician Norbert Wiener, who developed the rigorous mathematical framework underlying the phenomenon.
The origins of Brownian motion trace back to observations made by botanist Robert Brown in 1827, who noticed that pollen particles suspended in water appeared to move randomly. The eventual explanation emerged from molecular physics: countless microscopic collisions generated seemingly erratic motion; the Wiener process formalises this behaviour mathematically.
It possesses several defining characteristics:
changes occur continuously through time
increments are independent
increments are normally distributed
the expected change over any interval is zero
variance increases proportionally with time
The resulting path appears highly irregular despite being continuous; at any moment, movement may occur upward or downward, no direction is favoured. As such, the process exhibits no memory, meaning future movements are independent of past movements. This property makes the Wiener process a natural mathematical representation of uncertainty when countless small influences combine to generate observable outcomes.
Wiener Processes in Finance
The Wiener process occupies a central role within modern financial theory. With the geometric Brownian motion model used in the Black-Scholes framework assumes that asset prices evolve through a continuous diffusion process driven by an underlying Wiener process.
Under this approach, returns arise from two components:
a deterministic trend component
a stochastic component driven by Brownian motion
The result is a model in which prices move continuously and uncertainty accumulates gradually through time. The mathematical elegance of the Wiener process makes it highly attractive; analytical solutions often exist, calculations remain tractable, and many theoretical results can be derived explicitly. However, empirical market behaviour reveals important limitations. Financial markets frequently experience jumps, gaps, and extreme events that cannot be fully explained by continuous diffusion alone. This observation motivates the introduction of additional processes capable of representing discontinuities.
The Poisson Process
Where the Wiener process models continuous randomness, the Poisson process models random events.
Named after French mathematician Siméon Denis Poisson, the process describes situations in which events occur independently and randomly through time at an average rate.
Examples include:
telephone calls arriving at a call centre
customers entering a shop
equipment failures
earthquakes
insurance claims
corporate defaults
The defining feature is that events occur discretely rather than continuously; between events, nothing happens, then an event suddenly occurs. The process is characterised by a single parameter known as the arrival rate, often denoted by λ (lambda), which represents the average number of events expected over a given time period. Although the timing of individual events remains unpredictable, the overall statistical behaviour becomes highly predictable over long horizons.
The Logic of Random Arrivals
One of the most powerful features of the Poisson process is its simplicity.
If events occur independently and at a constant average rate, the probability of observing a specific number of events during a given period can be calculated directly. This allows researchers to model systems where uncertainty concerns not the magnitude of movement but the occurrence of events themselves. Importantly, the Poisson process also possesses a memoryless property; the probability of an event occurring in the future depends only on the future interval being considered and not on how much time has already elapsed. This feature makes Poisson models particularly useful when analysing phenomena where events emerge independently of previous occurrences.
Poisson Processes in Financial Markets
Although financial prices often move continuously, many important market events occur discretely.
Examples include:
earnings announcements
credit defaults
regulatory interventions
dvidend declarations
mergers and acquisitions
geopolitical shocks
sudden liquidity disruptions
These events frequently produce abrupt price adjustments that cannot be represented effectively through Brownian motion alone. A Poisson process provides a natural framework for modelling such arrivals; as rather than assuming information enters markets smoothly, the Poisson perspective recognises that significant information often arrives in bursts. Markets may remain relatively calm before suddenly repricing in response to a discrete event. Accordingly, this distinction becomes particularly important when analysing tail risk and systemic vulnerability.
Combining Wiener and Poisson Processes
Many modern financial models combine both processes, and the resulting frameworks acknowledge that markets exhibit two distinct forms of uncertainty:
continuous fluctuations generated by countless small influences
discrete jumps generated by major events
This combination produces a more realistic representation of observed market behaviour. Jump-diffusion models, first popularised by Robert Merton, incorporate a Wiener component to capture ordinary market noise and a Poisson component to capture sudden discontinuities; as under such models, asset prices evolve continuously most of the time but occasionally experience abrupt jumps. This structure aligns more closely with empirical evidence, where markets typically move gradually yet periodically experience dramatic repricing events.
Risk, Tail Events, and Model Limitations
The distinction between Wiener and Poisson processes carries significant implications for risk management.
A pure Wiener framework tends to underestimate the likelihood of extreme outcomes because all movements emerge from continuous diffusion. Large changes become increasingly improbable as their magnitude grows. On the other hand, a Poisson framework introduces the possibility that significant events may occur suddenly and without warning. The consequences are substantial, as risk estimates based solely on Brownian motion often produce overly optimistic assessments of portfolio resilience. Liquidity crises, credit events, and market crashes frequently arise through mechanisms more consistent with jump behaviour than continuous diffusion. The inclusion of Poisson-driven events therefore represents an important step toward more realistic risk modelling.
Nevertheless, neither process fully captures the complexity of real markets. Financial systems exhibit feedback loops, behavioural contagion, volatility clustering, adaptive behaviour, and regime shifts that extend beyond the assumptions of classical stochastic models. Thus, the challenge is not merely selecting the correct process but recognising that all models are simplifications of a far more intricate reality.
From Continuous Noise to Discrete Shocks
Viewed conceptually, the Wiener and Poisson processes represent two complementary ways of understanding uncertainty.
The Wiener process describes a world dominated by incremental change; small influences accumulate continuously, gradually shaping outcomes through time. The Poisson process describes a world shaped by sudden events; long periods of stability may be interrupted by discrete occurrences that alter the trajectory of the system. Importantly, real financial markets contain both; daily price fluctuations often resemble diffusion-like behaviour; major crises, policy announcements, and systemic shocks often resemble jump processes.
Understanding both mechanisms allows investors to develop a more nuanced view of market dynamics and a deeper appreciation for the diverse sources of uncertainty that influence asset prices.
The MorMag Perspective
At MorMag, uncertainty is viewed as multidimensional rather than singular.
Traditional financial models often focus heavily on continuous volatility while underappreciating the role of discontinuities. Yet history demonstrates that many of the most consequential market outcomes emerge not from ordinary fluctuations but from discrete events that fundamentally alter expectations, liquidity conditions, or market structure.
The Wiener process provides an elegant framework for understanding the accumulation of everyday uncertainty, likewise, the Poisson process offers an equally valuable framework for understanding the arrival of unexpected information and systemic shocks. Neither should be viewed as a complete description of reality. Instead, they represent foundational building blocks within a broader framework for understanding complex adaptive systems.
Effective capital allocation requires recognising both forms of uncertainty. Investors must navigate not only the continuous noise of markets but also the possibility of abrupt discontinuities that reshape the opportunity set. Robust decision-making therefore depends on understanding how diffusion, jumps, and structural change interact across time.
Conclusion
The Wiener and Poisson processes occupy a foundational position within modern probability theory and financial mathematics because they capture two fundamentally different forms of randomness.
The Wiener process models continuous uncertainty through the accumulation of countless small fluctuations; and the Poisson process models discrete uncertainty through the random arrival of events. Together, they provide the mathematical foundations for many of the models used to analyse asset prices, derivatives, risk, credit events, and market behaviour.
More importantly, they offer a deeper conceptual lesson: markets are shaped by both gradual evolution and sudden transformation. Understanding uncertainty therefore requires recognising not only the steady flow of everyday randomness but also the unpredictable arrival of events capable of changing the course of financial systems in an instant.
