Mathematics
The Poisson Distribution and Modeling Rare Events
Quick fact
The Poisson distribution was introduced by Siméon Denis Poisson in 1837, but it became famous in 1898 when Ladislaus Bortkiewicz used it to model the number of Prussian soldiers killed by horse kicks each year!
Why this is interesting
You've probably wondered why some days you get three phone calls in five minutes, and other days none at all. What if you could predict exactly how likely that is?
Read the full explanation
Understanding The Poisson Distribution and Modeling Rare Events
Imagine you are counting rare events, like meteor sightings in an hour or typos on a page. The Poisson distribution tells you the probability of seeing exactly k events when the average rate is λ. It's like a magic formula that uses only the average to give a whole probability picture. For example, if a call center gets an average of 5 calls per minute, the Poisson distribution can tell you the chance of getting 0, 1, 2, or even 10 calls in a minute. The key assumptions are that events happen independently and at a constant average rate—think of a steady drip of water where each drop falls randomly but at a fixed average pace. This distribution is perfect for rare events, where the number of trials is large but the probability of success is small, like counting defective products in a factory or radioactive decays in a sample.
A deeper explanation
The Poisson distribution arises from taking the binomial distribution to its limit: as the number of trials n goes to infinity and the success probability p goes to zero, with the product np equal to a constant λ, the binomial probabilities converge to the Poisson probabilities. This is called the law of rare events. The probability mass function is P(X=k) = (λ^k e^{-λ}) / k!, where λ is both the mean and the variance. This elegant property makes it easy to model: you only need to know the average rate. The Poisson process, which generates these counts, has inter-arrival times that follow an exponential distribution, a continuous analogue. This concept is foundational in queuing theory, reliability engineering, and even modeling cosmic ray hits. Its importance lies in turning a seemingly unpredictable phenomenon into a precise, computable tool, allowing scientists and engineers to make informed decisions under uncertainty.