Negative hyper geometric distribution: In sampling without replacement from the population comprising of r elements of one kind and N - r of another, if two elements corresponding to which selected are replaced every time, then the probability of finding x elements of first kind in a random sample of n elements can be given as follows
The mean of distribution can be given as Nr/N and the variance can be given as follows (nr/N)(1-r/N)(N+n)/(N+1).
It corresponds to the beta binomial distribution with the integral parameter values.