Discrete Probability Distributions: Binomial and Poisson Discrete probability distributions are a powerful tool for analyzing situations where there are...
Discrete probability distributions are a powerful tool for analyzing situations where there are a finite number of possible outcomes and the outcome of each trial is independent of other trials. These distributions help us calculate the probability of specific outcomes in a sequence of independent events.
Binomial distribution:
Imagine flipping a coin twice. Each flip represents an independent event with two possible outcomes: heads (H) or tails (T).
The binomial distribution helps us calculate the probability of getting a specific number of heads or tails in a sequence of n coin flips.
It can be used to model situations with a limited number of trials and a finite number of possible outcomes.
Poisson distribution:
Imagine throwing a dart at a dartboard. Each throw represents an independent event where the dart lands on a specific point on the board.
The Poisson distribution helps us calculate the probability of a specific number of darts landing in a specific area or number of darts hitting a specific point on the board.
It can be used to model situations where the number of events that occur in a short period of time follows a predictable pattern.
Key differences between binomial and Poisson:
| Feature | Binomial | Poisson |
|---|---|---|
| Number of trials | Finite | Infinite |
| Possible outcomes per trial | 2 (heads/tails) | 1 (land on specific point) |
| Probability calculation | Probability of getting a specific number of successes in n independent trials | Probability of a specific number of events occurring in a fixed amount of time |
| Applications | Coin flips, choosing a sample, modeling a limited number of successes in a sequence of independent events | Dart throwing, modeling continuous random variables, modelling a Poisson process (rare events happening at a constant rate) |
Important properties:
Both distributions are discrete, meaning the number of outcomes is finite.
The sum of independent binomial or Poisson random variables is also a binomial or Poisson random variable, respectively.
The expected value for both distributions is equal to the number of trials multiplied by the probability of the most likely outcome.
Examples:
Rolling a die is an example of the binomial distribution, with 6 possible outcomes and each trial being independent.
Counting the number of accidents on a busy highway follows a Poisson distribution, with a constant rate of 5 accidents per hour