A discrete random variable X is said to have negative binomial distribution if its probability density function is defined as,
where r>0 and 0<=p<=1 are the two parameter of the distribution such that p+q=1. f(x;r,p) is the probability of getting exactly r successes in (x+r) independent bernoulli trials.
Properties of Negative Binomial Distribution
- The mean of negative binomial distribution is .
- The variance of the distribution is.
- The skewness of the distributionis.
- The kurtosis of negative binomial distribution is .
Characteristics of Negative Binomial Distribution
- If r=1, negative binomial distribution tends to geometric distribution with parameter p.
- If r➝∞, q➝0 and rq➝λ, then negative binomial distribution tends to poission distribution.
- The variance of negative binomial distribution is always greater than mean.
- The distribution is positively skewed and leptokurtic.
Applications of Negative Binomial Distribution
- The negative binomial distribution is applicable in those data set where variance is greater than mean.
- When poission unable to describe a data set or inadequate then we prefer negative binomial distribution.
- Used in accident statistics ( birth and death process).
- Used in psychological data set.
- Also used in time series data in economics, medical and military data etc.