N = 20, K = 8, n = 5, k = 2
C(8,2)C(12,3)/C(20,5)
0.3973.
Drawing items without replacement from , of which are special, the number of special items drawn has , mean and variance .
This page is easier if you are comfortable with:
A batch of has defective items; we inspect . and . The mean is and the variance . The binomial coefficients are the counts of binomial coefficients and combinations.
Unlike the binomial distribution (the binomial distribution), the trials are dependent, because each draw changes the composition. The factor in the variance (the finite population correction) shrinks it relative to the binomial value . When is large compared with , the hypergeometric is close to the binomial with .
Applications include quality control, lotteries (matching numbers), card hands and capture–recapture estimates (capture–recapture). The counting argument is the same as for the multiplication rule and the counting product rule.
The hypergeometric distribution gives exact probabilities when sampling without replacement.
Most samples contain one or no defective items.
These examples are worked by hand, step by step, so you can compare your own working.
Find for , , .
Find .
Find the variance of .
A few situations come up again and again. Knowing them saves time and helps you spot answers that cannot be right.
C(8,2)C(12,3)/C(20,5)
0.3973.
nK/N
2.
n(K/N)(1 − K/N)(N − n)/(N − 1)
0.9474.
(N − n)/(N − 1)
15/19 = 0.7895.
Drawing n items without replacement from N with K special: P(X = k) = C(K, k)C(N − K, n − k)/C(N, n), mean nK/N, variance n(K/N)(1 − K/N)(N − n)/(N − 1).
Find P(X = 2) for N = 20, K = 8, n = 5.
Find nK/N.
Find the variance.
Find the finite population factor.
independent draws ✗
hypergeometric for no replacement ✓
Probabilities change.
npq ✗
npq(N − n)/(N − 1) ✓
Smaller variance.
C(N, k) ✗
C(N, n) ✓
Total samples.
allowed ✗
k ≤ min(n, K) ✓
Support.
Try each one on paper, then open the answer. Use the solver above to see the full working.
28
220
15504
4
0.5
0.0333
Unlimited fresh questions with instant marking. Choose a level, work it out on paper, then type your answer.
Solve: …
The count of special items when sampling without replacement.
nK/N.
Draws are not independent.
When N is much larger than n.
binomial coefficient.
More in Advanced statistics and stochastic processes and the formula sheets.