The Poisson Distribution is typically used to model the number of events that happen in a certain time period, especially where the expected number of events is very small.
The Poisson distribution is parameterized simply by $\lambda$, which can be thought of as the long-run average number of events in the time period of interest. To illustrate, if a device manufacturer on average produces 10 defective parts per day, we can model the expected number of defective parts on any given day as $X \sim Poisson(\lambda=10)$.
The Poisson is a limiting case of the [[Binomial distribution]] when $n$ gets really large and $p$ gets really small.
Some common examples include
- Number of winning lottery tickets sold in a given time period.
- Number of animals captured in a trap in a given time period.
- Number of vehicles crossing a bridge in one day
## Notation
$X \sim Poisson(\lambda) $
## Probability Mass Function
$P(X=x) = \frac{\lambda^x}{x!}e^{-\lambda}$
## Expectation
$E(X) = \lambda$
## Variance
$V(X) = \lambda$
## R notation
The Poisson distribution is `pois` in R.
```R
# Probability mass function
prob <- dpois(x, lambda)
# Cumulative distribution function
cum_prob <- ppois(q, lambda)
# Quantile function
quantile_val <- qpois(p, lambda)
# Random number generation
random_values <- rpois(n, lambda)
```