You are on page 1of 1

These are similar in principle to the binomial distribution, with one major exception the hypergeometric probability is for

situations where you samplewithout replacement, and the binomial distribution assumes sampling with replacement. The general formula is as follows: N = size of population M = # of items in population with property "E" N-M = # of items in population without property "E" n = number of items sampled k = number of items in sample with property "E" P(k) = (M,k)*(N-M,n-k) / (N,n) In general, the "symbol" (a,b) means combinations of "a items, taking b of them:" (a,b) = aCb = a! / (b!(a-b)!) Example - The Lottery 48 numbered balls are inside a spinning container, and 6 are drawn at random from the container, representing the winning combination. What is the probability of matching at least 4 of the combination of 6? N = 48 M=6 n=6 k = 4,5,6 P(k) = P(6) = P(5) = P(4) = (M,k)*(N-M,n-k) / (N,n) (6,6)*(42,0) / (48,6) = 1 * 1 / 12271512 = .00000008 (6,5)*(42,1) / (48,6) = 6 * 42 / 12271512 = .000021 (6,4)*(42,2) / (48,6) = 15 * 861 / 12271512 = .00105

P(at least 4) = P(4) + P(5) + P(6) = .00107 Example - Pulling Marbles from a Box w/o Replacement A box contains 20 marbles, 5 of which are red, the remainder are white. If you draw 3 at random from the box, what is the probability of having at least 2 red marbles? N = 20 M=5 n=3 k = 2,3 P(k) = (M,k)*(N-M,n-k) / (N,n) P(2) = (5,2)*(15,1) / (20,3) = .1316 P(3) = (5,3)*(15,0) / (20,3) = .0088 P(at least 2) = P(2) + P(3) = .1316 + .0088 = .1404

You might also like