Applications of Binomial Distribution

182 Binomial and Poisson Distributions
This is the general formula for the binomial probability distribution. We shall see how to apply
this formula in the next section, and you will find that it is not quite as fearsome as it may look
at first.
B. APPLICATIONS OF THE BINOMIAL DISTRIBUTION

Example 1:
The probability that a match will break on being struck is 0.04. What is the probability that,
out of a box of 50:
(a) none will break;
(b) more than 2 will break?
A match will either break or not break when it is struck. Therefore:
P(breaking) = 0.04 = P(success) = p
P(not breaking) = 1 − p = 1 − 0.04 = 0.96 = P(failure)
We have a box of 50 matches, so n = 50.
(a) We require the probability that none will break, i.e. the probability of no successes,
P(0).
P(0) = nC0p0(1 − p)n–0 using general formula:
n!
= (1 − p)n = (1 − p)n
n! 0!
= (0.96)50 = 0.1299 to 4 decimal places.
Therefore, probability none will break = 0.1299
(b) Probability that more than 2 will break = 1 − probability that 2 or less will break.
Probability that 2 or less will break = probability that 0 or 1 or 2 will break
= P(0) + P(1) + P(2)
We thus need to work out P(1) and P(2):
P(1) = nC1p1(1 − p)n–1
n! 50!
= p(1 − p)n−1 = (0.04)(0.96) 49
(n − 1)! 49!
= 50 × (0.04)(0.96)49 = 0.2706 to 4 decimal places.
P(2) = nC2p2(1 − p)n–2
n! 50!
= p 2 (1 − p)n-2 = (0.04) 2 (0.96) 48
(n − 2)!2! 48! 2!
50 × 49
= (0.04) 2 (0.96) 48 = 0.2762
2
Therefore, probability that more than 2 will break:
= 1 − (0.1299 + 0.2706 + 0.2762)
= 1 − 0.6767 = 0.3233
= 0.323 to 3 decimal places.
© ABE
Binomial and Poisson Distributions 183
Example 2:
It has been found that, on average, 5% of the eggs supplied to a supermarket are cracked. If
you buy a box of 6 eggs, what is the probability that it contains 2 or more cracked eggs?
An egg is either cracked or not cracked:
P(cracked) = 5% = 0.05 = P(success) = p
P(not cracked) = 1 − p = 1 − 0.05 = 0.95 = (failure)
We have a box of 6 eggs, so n = 6.
Probability of 2 or more cracked eggs in a box:
= 1 − probability of less than 2 cracked eggs in a box
= 1 − probability of 0 or 1 cracked eggs in a box
= 1 − [P(0) + P(1)].
P(0) = nC0p0(1 − p)n–0 = (1 − p)n = (0.95)6 = 0.7351
n!
P(1) = nC1p1(1 − p)n–1 = p(1 − p)n−1
(n − 1)! 1!
6!
= (0.05)(0.95)5 = 6(0.05)(0.95)5 = 0.2321
5!
Therefore, probability of 2 or more cracked eggs in a box:
= 1 − (0.7351 + 0.2321)
= 1 − 0.9672
= 0.0328 = 0.033 to 3 decimal places.
Example 3:
A retail sales manager will accept delivery of a large consignment of goods if a random
sample of 10 items contains no defectives. If 3% of the producer's total output is defective,
what is the probability that delivery of a consignment will be accepted? How would the
situation change if the random sample were of only 5 items?
An item is either defective or non-defective. Therefore:
P(defective) = 3% = 0.03 = P(success) = p
P(non-defective) = 1 − p = 1 − 0.03 = 0.97 = P(failure).
First, the manager takes a sample of 10, so n = 10.
We require the probability that this sample contains no defectives, i.e. P(0):
P(0) = nC0p0(1 − p)n–0 = (1 − p)n
= (0.97)10
= 0.7374 to 4 decimal places.
Therefore, probability that a delivery will be accepted is 0.7374
Secondly, consider a sample of 5.
P(0) = (1 − p)n = (0.97)5 = 0.8587 to 4 decimal places.
Therefore, probability that delivery will be accepted is 0.8587, which is higher than when a
larger sample was taken.
© ABE
184 Binomial and Poisson Distributions
Notes
1. One of the conditions for using the binomial distribution is that the chance of success,
p, must be constant throughout the series of trials. This means that if we are, say,
taking items from a batch and not replacing them before the next item is taken, then the
binomial distribution does not apply because the batch is fractionally smaller (by 1 item
each time). In practice, however, when the batch from which a sample is being taken is
very large compared with the sample, the binomial distribution is a satisfactory
approximation. As a rough guide, you can consider the batch to be very large if it is
more than about 10 times the sample size.
2. Tables are available giving values of nCx for various values of n and x. This is
particularly useful for large values of n but in examinations usually you are expected to
be able to work them out for yourself. Most calculators have keys for calculating the
number of combinations and the number of permutations.
C. MEAN AND STANDARD DEVIATION OF THE BINOMIAL

DISTRIBUTION
In the case of a theoretical distribution like the binomial distribution there is usually a simple
formula to show what the mean and standard deviation ought to be without needing to go
through any lengthy calculations.
For any binomial distribution, the mean µ and the standard deviation σ are given by the
following formulae:
µ = np
σ = npq = np(1 − p)
1 5
Thus, if n = 3, p = ,1− p =
6 6
therefore:
1 1
µ = 3× = = 0.5
6 2
1 5 15
σ = 3× × = = 0.645 to 3 significant figures.
6 6 36
What we are saying is that if we repeat this experiment very many times, the mean number
of successes is 0.5. As with a frequency distribution, the mean does not necessarily have to
be one of the original values of x.
D. THE POISSON DISTRIBUTION

Introduction
The binomial distribution is useful in cases where we take a fixed sample size, and count the
number of successes. Sometimes we do not have a definite sample size, and then the
binomial distribution is of no use. For example, if we are studying the occurrence of
accidents, we can count how many accidents occur in a month, but it is nonsense to talk
about how many accidents did not occur! In such cases we use another theoretical
distribution called the Poisson distribution (named after the French mathematician).
© ABE

Applications of Binomial Distribution

Uploaded by

Copyright:

Available Formats

Applications of Binomial Distribution

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applications of Binomial Distribution

Uploaded by

Copyright:

Available Formats

182 Binomial and Poisson Distributions

B. APPLICATIONS OF THE BINOMIAL DISTRIBUTION

C. MEAN AND STANDARD DEVIATION OF THE BINOMIAL

D. THE POISSON DISTRIBUTION

You might also like