Bernoulli Distribution
An r.v. is said to have the Bernoulli distribution with parameter if and , where . We write this as . (or , as a special case of binomial distribution)
Bernoulli Trial
An experiment that can result in either a "success" of a "failure" (but not both) is called Bernoulli trial. A Bernoulli random variable can be thought of as the indicator of success in a Bernoulli trial: it equals if success occurs and if failure occurs in the trail.
Binomial Distribution
Suppose that independent Bernoulli trials are performed, each with the same success probability . Let be the number of successes. The distribution of is called the Binomial distribution with parameters and . We write or to mean that has the Binomial distribution with parameters is , where is a positive integer and
Properties
If , then the PMF of is
for , with and
Let , and , then
Let with and even. Then the distribution of is symmetric about , that is, for all nonnegative integers
Regenerative Property
Let and , if and are independent, we have .
To generalize, if and are independent, we can conclude
Negative Binomial Distribution (Pascal Distribution)
This distribution models the number of failures in a sequence of independent and identically distributed Bernoulli trials before a specified (non-random) number of successes (denoted ) occurs.
The PMF of the negative binomial distribution is
and