What is the Poisson distribution?
The Poisson distribution counts events in a fixed interval when they happen at a known average rate and independently of the time since the last one. Its mean is λ and its standard deviation is √λ. If an interval is a different length from the one your average is for, scale λ first: 12 loaves per half hour is 2 per five minutes.
Source OpenStax Introductory Statistics 2e · 4.6 Poisson Distribution ↗
How to calculate Poisson probability
Here it is with your numbers:
- Exactly 0: e^−0.75 × 0.75^0 ÷ 0! = 0.4724
- Exactly 1: e^−0.75 × 0.75^1 ÷ 1! = 0.3543
- At most 1, adding those: 0.8266
- More than 1: 1 − 0.8266 = 0.1734
It means that over many intervals like this one, about 17.34% of them would have more than 1 event.
Source NIST/SEMATECH e-Handbook · 1.3.6.6.19 Poisson Distribution ↗
Poisson probability table
Each count with its chance and the running total, P(X ≤ k), around the average. The counts your answer includes are highlighted.
| Events (k) | P(X = k) | P(X ≤ k) |
|---|---|---|
| 0 | 0.4724 | 0.4724 |
| 1 | 0.3543 | 0.8266 |
| 2 | 0.1329 | 0.9595 |
| 3 | 0.0332 | 0.9927 |
| 4 | 0.0062 | 0.9989 |
| 5 | 0.0009 | 0.9999 |
| 6 | 0.0001 | > 0.9999 |
| 7 | 1.2511 × 10⁻⁵ | > 0.9999 |
| 8 | 1.1729 × 10⁻⁶ | > 0.9999 |
| 9 | 9.7739 × 10⁻⁸ | > 0.9999 |
| 10 | 7.3304 × 10⁻⁹ | > 0.9999 |
| 11 | 4.998 × 10⁻¹⁰ | > 0.9999 |
| 12 | 3.1238 × 10⁻¹¹ | > 0.9999 |
| 13 | 1.8022 × 10⁻¹² | > 0.9999 |
| 14 | 9.6545 × 10⁻¹⁴ | > 0.9999 |
| 15 | 4.8272 × 10⁻¹⁵ | 1 |
| 16 | 2.2628 × 10⁻¹⁶ | 1 |
| 17 | 9.9828 × 10⁻¹⁸ | 1 |
| 18 | 4.1595 × 10⁻¹⁹ | 1 |
| 19 | 1.6419 × 10⁻²⁰ | 1 |
| 20 | 6.1572 × 10⁻²² | 1 |
| 21 | 2.199 × 10⁻²³ | 1 |
| 22 | 7.4965 × 10⁻²⁵ | 1 |
| 23 | 2.4445 × 10⁻²⁶ | 1 |
| 24 | 7.6391 × 10⁻²⁸ | 1 |
| 25 | 2.2917 × 10⁻²⁹ | 1 |
| 26 | 6.6108 × 10⁻³¹ | 1 |
| 27 | 1.8363 × 10⁻³² | 1 |
| 28 | 4.9187 × 10⁻³⁴ | 1 |
| 29 | 1.2721 × 10⁻³⁵ | 1 |
| 30 | 3.1802 × 10⁻³⁷ | 1 |
Poisson distribution examples
| Example (OpenStax) | Probability |
|---|---|
| More than 1 phone call in 15 minutes (average 0.75) | 0.1734 |
| Exactly 3 loaves in 5 minutes (average 2) | 0.1804 |
| Fewer than 5 bad checks in a day (average 6) | 0.2851 |
| At most 160 emails in a day (average 147) | 0.8666 |
| More than 2 “uh”s in a broadcast (average 2) | 0.3233 |
FAQs
What is Poisson distribution in simple words?
It gives the chance of a certain number of events in a fixed interval of time or space, when the events happen at a known average rate and independently of each other, like phone calls per hour or typos per page.
Source OpenStax Introductory Statistics 2e · 4.6 Poisson Distribution ↗
What is the formula for Poisson?
P(X = x) = e^−λ λ^x ÷ x!, where λ is the average number of events in the interval. Add these for x = 0 up to your number to get “at most”.
Source NIST/SEMATECH e-Handbook · 1.3.6.6.19 Poisson Distribution ↗
When should you use Poisson distribution?
When you are counting events in an interval (an hour, a page, a mile), you know the average count, and events don’t affect each other. It also approximates the binomial distribution when there are many trials and a small chance of success.
Source OpenStax Introductory Statistics 2e · 4.6 Poisson Distribution ↗
What are 5 real life examples of Poisson distribution?
OpenStax’s examples include loaves put on a bakery shelf, bad checks a bank receives, phone calls, emails a day and times a reporter says “uh”, each counted in a fixed interval with a known average.
Sources
- OpenStax Introductory Statistics 2e: 4.6 Poisson Distribution
- NIST/SEMATECH e-Handbook: 1.3.6.6.19 Poisson Distribution