Lecture 7: Special Probability Distributions - 2: Assist. Prof. Dr. Emel YAVUZ DUMAN

The Hypergeometric Distribution
The Poisson Distribution
Lecture 7: Special Probability Distributions - 2

Assist. Prof. Dr. Emel YAVUZ DUMAN
Introduction to Probability and Statistics
Istanbul K
ult
ur University
Outline
Outline
Many times we used sampling with and without replacement to

illustrate the multiplication rules for independent and dependent
events. To obtain a formula analogous to the binomial distribution
that applies to sampling without replacement, in which case the
trials are not independent, let us consider a set of N elements for
which M are looked upon as successes and the other N M as
failures. In connection with the binomial distribution, we are
interested in the probability of getting x successes in n trial, but
now we are choosing, without replacement, n of the N elements
contained in the set.
There are
M
x way of choosing x of the M successes, and
NM
ways of choosing n x of the N M failure, and
nx
M NM
hence x nx ways of choosing x successes and n x
failures.

Since there are Nn ways of choosing n of the N elements in
the set, and we shall assume that they are all equally likely
(which is what we mean when we say that the selection is
random),
N that the probability of x successes in n trials is
M NM then
x
nx / n .
Denition 1
A random variable X has a hypergeometric distribution and it is
referred to as a hypergeometric random variable if and only if its
probability distribution is given by
M NM
h(x; n, N, M) =
Nnx

n
for x = 0, 1, 2, , n, x M and n x N M.
Thus, for sampling without replacement, the number of successes
in n trials is a random variable having a hypergeometric
distribution with parameters n, N, and M.
Example 2
As part of an air-pollution survey, an inspector decides to examine
the exhaust of six of a companys 24 trucks. If four of the
companys trucks emit excessive amounts of pollutants, what is the
probability that none of them will be included in the inspectors
sample?
Solution. Substituting x = 0, n = 6, N = 24, and M = 4 into the
formula for the hypergeometric distribution, we get
4244
M NM
h(x; n, N, M) = h(0; 6, 24, 4) =
Nnx

n
2460

6
= 0.2880.
Example 3
Draw 6 cards from a deck without replacement. What is the
probability of getting two hearts?
Solution. Substituting x = 2, n = 6, N = 52, and M = 13 into
the formula for the hypergeometric distribution, we get
135213
M NM
h(x; n, N, M) = h(2; 6, 52, 13) =
Nnx

n
5262

6
= 0.31513.
Example 4
49 balls are numbered 1 - 49. You select six numbers between 1
and 49. The ones you write on your lotto card. What is the
probability that they contain (a) match 4? (b) match 6?
Solution. (a) Substituting x = 4, n = 6, N = 49, and M = 6 into
the formula for the hypergeometric distribution, we get
6496
M NM
h(x; n, N, M) = h(4; 6, 49, 6) =
Nnx
4964
= 2.3062105 .
(b) Substituting x = 6, n = 6, N = 49, and M = 6 into the

formula for the hypergeometric distribution, we get
6496
M NM
h(x; n, N, M) = h(6; 6, 49, 6) =
nx
N
n
4966

6
= 7.1511108 .
Theorem 5
The mean and the variance of the hypergeometric distribution are
=
nM(N M)(N n)
nM
and 2 =
.
N
N 2 (N 1)
Example 6
Suppose that a researcher goes to a small college of 200 faculty, 12
of which have blood type O-negative. She obtains a simple random
sample of 20 of the faculty. Determine the mean and standard
deviation of the number of randomly selected faculty that will have
blood type O-negative.
Solution. Substituting n = 20, N = 200, and M = 12 into the
formula for the hypergeometric distributions mean and variance
we obtain
20 12
nM
=
= 1.2
=
N
200
and

nM(N M)(N n)
20 12(200 12)(200 20)
=
= 1.0101
=
2
N (N 1)
2002 (200 1)
We expect that, in a random sample of 20 faculty members, 1.2
will have blood type O-negative.
Example 7
A case of wine has 12 bottles, 3 of which contains spoiled wine. A
sample of 4 bottles is randomly selected from the case.
(a) Find the probability distribution for X , the number of spoiled
wine in the sample
(b) What are the mean and variance of X ?
Solution. For this example n = 4, N = 12, and M = 3. Then
3 9
h(x; 4, 12, 3) =
124x
.
4
h(x; 4, 12, 3) =

9
(x3)(4x
)
.
12
(4)
(a) The possible values for X are 0, 1, 2 and 3, with probabilities

39
39
h(0; 4, 12, 3) = 0124 = 0.25, h(1; 4, 12, 3) = 1123 = 0.51,
4
39
39
h(2; 4, 12, 3) = 2122 = 0.22, h(3; 4, 12, 3) = 3121 = 0.02.

4
(b) The mean is given by

=
43
nM
=
=1
N
12
and the variance is

2 =
4 3(12 3)(12 4)
nM(N M)(N n)
=
= 0.5455.
N 2 (N 1)
122 (12 1)
Binomial Approximation to Hypergeometric Distribution
When N is large and n is relatively small compared to N (the usual

rule of thumb is that n should not exceed 5 percent of N), there is
not much dierence between sampling with replacement and
sampling without replacement, and the formula for the binomial
distribution with the parameters n and = M
N may be used to
approximate hypergeometric probabilities.
Example 8
Among the 120 applicants for a job, only 80 are actually qualied.
If ve of the applicants are randomly selected for an in-depth
interview, nd the probability that only two of the ve will be
qualied for the job by using
(a) the formula for the hypergeometric distribution;
(b) the formula for the binomial distribution with = 80/120 as
an approximation.
Solution. (a) Substituting x = 2, n = 5, N = 120, and M = 80

into the formula for the hypergeometric distribution, we get
8040
h(x; n, N, M) = h(2; 5, 120, 80) =
1203
= 0.164.
rounded to three decimals;

(b) substituting x = 2, n = 5, N = 120, and =
formula for the binomial distribution, we get

2
b 2; 5,
3
80
120
2
3
into the
2

5
2
2 3
=
= 0.165
1
2
3
3
rounded to three decimals. As can be seen from these results, the

approximation is very close.
Example 9
Boxes contain 2000 items of which 10% are defective. Find the
probability that no more than 2 defectives will be obtained in a
sample of size 10.
Solution. For this question x is equal to 0, 1 or 2, n = 10,
N = 2000 and M = 2000 0.10 = 200. Since
n = 10 100 = 2000 0.05 = N 0.05
this means n is not exceed 5 percent of N we may use the method
of binomial approximation to the hypergeometric distribution also.
(a) The hypergeometric distribution:

P(X 2) = P(X = 0) + P(X = 1) + P(X = 2)
2001800 2001800 2001800
=
200010
10
20009
10
20008
10
= 0.3476 + 0.3881 + 0.1939 = 0.9296.

(b) Binomial approximation to the hypergeometric distribution
with = M/N = 200/2000 = 0.1:
P(X 2) = P(X = 0) + P(X = 1) + P(X = 2)

10
10
10
0
1
1
9
0.1 0.9 +
0.12 0.98
=
0.1 0.9 0 +
1
2
0
= 0.3487 + 0.3874 + 0.1937 = 0.9298.
Outline
When n, the number of trial, is large the calculation of binomial

probabilities with the formula of binomial distribution will usually
involve a prohibitive amount of work. In this section we shall
present a probability distribution that can be used to approximate
binomial probabilities of this kind. Specically, we shall investigate
the limiting form of the binomial distribution when n , 0,
while n remains constant.
Denition 10
A random variable X has a Poisson distribution and it is referred
to as Poisson random variable if and only if its probability
distribution is given by
p(x; ) =
x e
for x = 0, 1, 2
x!
where , the mean number of successes.

In general, Poisson distribution will provide a good approximation
to binomial probabilities when n 20 and 0.05. When
n 100 and n < 10, the approximation will generally be excellent.
Example 11
If 2 percent of books bound at a certain bindery have defective
bindings, use the Poisson approximation to the binomial
distribution to determine the probability that ve of 400 books
bound by this bindery will have defective bindings.
Solution. Substituting x = 5, = n = 400 0.02 = 8 into the
formula for Poisson distribution, we get
p(5; 8) =
85 e 8
= 0.09160.
5!
Example 12
Records show that the probability is 0.00005 that a car will have a
at tire while crossing a certain bridge. Use the Poisson
distribution to approximate the binomial probabilities that, among
10,000 cars crossing the bridge
(a) exactly two will have a at tire;
(b) at most two will have a at tire.
Solution. (a) Substituting x = 2,
= n = 10, 000 0.00005 = 0.5 into the formula for Poisson
distribution, we get
p(2; 0.5) =
0.52 e 0.5
= 0.07582.
2!
(b)
0.52 e 0.5 0.51 e 0.5 0.50 e 0.5
+
+
2!
1!
0!
= 0.07582 + 0.30327 + 0.60653 = 0.98562.
p(2; 0.5) + p(1; 0.5)+p(0; 0.5) =
Having derived the Poisson distribution as a limiting form of the

binomial distribution, we can obtain formulas for its mean and its
variance by applying the same limiting conditions (n , 0
and n = remains constant) to mean and the variance of the
binomial distribution. For the mean we get = n = and for the
variance we get 2 = n(1 ) = (1 ) which approaches
when 0.
Theorem 13
The mean and the variance of the Poisson distribution are given by
= and 2 = .
Theorem 14
The moment generating function of the Poisson distribution is
given by
t
MX (t) = e (e 1) .
Although the Poisson distribution has been derived as a limiting

form of the binomial distribution, it has many applications that
have no direct connection with binomial distribution.
In many practical situations we are interested in measuring how
many times a certain event occurs in a specic time interval or in a
specic length or area. For instance:
1
the number of phone calls received at an exchange or call

center in an hour;
the number of customers arriving at a toll booth per day;
the number of aws on a length of cable;
the number of cars passing using a stretch of road during a

day.
The Poisson distribution plays a key role in modeling such

problems.
Suppose we are given an interval (this could be time, length, area

or volume) and we are interested in the number of successes in
that interval. Assume that the interval can be divided into very
small subintervals such that:
1
the probability of more than one success in any subinterval is

zero;
the probability of one success in a subinterval is constant for

all subintervals and is proportional to its length;
subintervals are independent of each other.
We assume the following.

1
The random variable X denotes the number of successes in

the whole interval.
is the mean number of successes in the interval.
X has a Poisson Distribution with parameter and

P(X = x) = p(x; ) =
x e
, x = 0, 1, 2, .
x!
Example 15
The average number of trucks on any one day at a truck depot in
a certain city is known to be 12. What is the probability that on a
given day fewer than nine trucks will arrive at this depot?
Solution. Let X be the number of trucks arriving on a given day.
Then, using Poisson distribution with = 12, we get
P(X < 9) =
8

p(x; 12) =
x=0
8

12x e 12
x=0
121
x!
122
123 124
120
+
+
+
+
0!
1!
2!
3!
4!

5
6
7
8
12
12
12
12
+
+
+
+
5!
6!
7!
8!
= 0.1550.
= e 12
Example 16
The number of aws in a ber optic cable follows a Poisson
distribution. The average number of aws in 50m of cable is 1.2.
(a) What is the probability of exactly three aws in 150m of
cable?
(b) What is the probability of at least two aws in 100m of cable?
(c) What is the probability of exactly one aw in the rst 50m of
cable and exactly one aw in the second 50m of cable?
Solution. (a) Mean number of aws in 150m of cable is
1.2 3 = 3.6. So the probability of exactly three aws in 150m of
cable is
3.63 e 3.6
= 0.21247
p(3; 3.6) =
3!
(b) Mean number of aws in 100m of cable is 1.2 2 = 2.4. Let X

be the number of aws in 100m of cable.
P(X 2) = 1 P(X < 2) = 1 (P(X = 0) + P(X = 1))
= 1 p(0; 2.4) p(1; 2.4)
2.40 e 2.4 2.41 e 2.4
0!
1!
= 0.69156
=1
(c) Now let X denote the number of aws in a 50m section of

cable. Then we know that
1.21 e 1.2
= 0.36143.
P(X = 1) = p(1; 1.2) =
1!
As X follows a Poisson distribution, the occurrence of aws in the
rst and second 50m of cable are independent. Thus the
probability of exactly one aw in the rst 50m and exactly one aw
in the second 50m is
(0.36143)(0.36143) = 0.13063.
Example 17
Births in a hospital occur randomly at an average rate of 1.8 births
per hour. What is the probability of observing 4 births in a given
hour at the hospital?
Solution. If we let X be the number of births in an hour, then X
has a Poisson distribution:
P(X = 4) = p(4; 1.8) =
1.84 e 1.8
= 0.072302.
4!
Example 18
Consider a telephone operator who, on the average, handles ve
calls every 3 minutes. (a) What is the probability that there will
be no calls in the next minute? (b) At least one call?
Solution. If we let X be the number of calls in a minute, then X
has a Poisson distribution with = 53 . So
(a) P(no calls in the next minute) = P(X = 0) = p(0; 5/3) =
(5/3)0 e 5/3
= 0.1889
0!
(b) P(At least one call) = P(X 1) = 1 P(X = 0) =
1 p(0; 5/3) = 1 0.1889 = 0.8111.
Example 19
A certain kind of sheet metal has on the average, ve defects per
10-square-feet. If we assume a Poisson distribution, what is the
probability that a 15-square-feet sheet of the metal will have at
least six defects?
Solution. Let X denote the number of defects in a 15-square-foot
sheet of the metal. Then, since the unit area is 10-square-feet, we
have
= 5 1.5 = 7.5
and
P(X 6) = 1 P(X 5) = 1 (P(X = 0) + P(X = 1) + P(X = 2)
+P(X = 3) + P(X = 4) + P(X = 5))

0
7.51
7.52 7.53
7.54
7.55
7.5
+
+
+
+
+
= 1 e 7.5
0!
1!
2!
3!
4!
5!
= 1 (0.2414)
= 0.7586.
Thank You!!!

Lecture 7: Special Probability Distributions - 2: Assist. Prof. Dr. Emel YAVUZ DUMAN

Uploaded by

Copyright:

Available Formats

Lecture 7: Special Probability Distributions - 2: Assist. Prof. Dr. Emel YAVUZ DUMAN

Uploaded by

Document Information

Original Description:

Original Title

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

Lecture 7: Special Probability Distributions - 2: Assist. Prof. Dr. Emel YAVUZ DUMAN

Uploaded by

Copyright:

Available Formats

The Hypergeometric Distribution

The Poisson Distribution