DISTRIBUTIONS RELATED TO THE

NORMAL DISTRIBUTION

Gabriel Asare Okyere (Ph.D.)

October 18, 2016

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 1 / 38
Outline of Presentation

1 INTRODUCTION

2 SAMPLING DISTRIBUTION OF MEANS

3 THE CENTRAL LIMIT THEOREM

4 SAMPLING DISTRIBUTION OF PROPORTIONS

5 SAMPLING DISTRIBUTION OF THE VARIANCE

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 2 / 38
INTRODUCTION

The idea that sample observations are random variables

presupposes that any quantity computed from sample
observations can also be a random variable.

As a random variable, the computed statistic has a probability

distribution of its own called probability distribution of the
statistic or sampling distribution of the statistic.

For instance, if X1 , X2 , ..., Xn is a random sample of size n, then

the sample mean x̄, the sample variance s 2 or the sample
standard deviation s which are statistics that could be computed
from the sample also have their own probability distributions.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 3 / 38
INTRODUCTION

Some of the sampling distributions to be consider include

Sampling distribution of x̄

Sampling distribution of p̂

Sampling distribution of s 2

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 4 / 38
Outline of Presentation

1 INTRODUCTION

2 SAMPLING DISTRIBUTION OF MEANS

3 THE CENTRAL LIMIT THEOREM

4 SAMPLING DISTRIBUTION OF PROPORTIONS

5 SAMPLING DISTRIBUTION OF THE VARIANCE

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 5 / 38
SAMPLING DISTRIBUTION OF MEANS

Let x1 , x2 , ..., xn be the values assumed by a corresponding set of

random variables X1 , X2 , ..., Xn ; we refer to the xi ’s as a random
sample of size n.

We can recall that X̄ is a specific value that can be calculated

from a specific sample. If we consider only one sample, this is
correct, however if we consider more than one sample, we need
to think of X̄ differently. The value of X̄ usually changes as the
sample changes.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 6 / 38
SAMPLING DISTRIBUTION OF MEANS
The mean X̄ of such a sample is a value assumed by the random
variable
X1 + X2 + ... + Xn
X̄ =
n

Therefore X̄ is a random variable since X1 , X2 , ..., Xn are random

variables corresponding to n trials of the sample values
x1 , x2 , ..., xn .

DEFINITION
The probability distribution of the statistic X̄ is called the sampling
distribution for the sample mean or the sampling distribution of
means.
Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION
October 18, 2016 7 / 38
SAMPLING DISTRIBUTION OF MEANS
For instance, suppose a population consists of four members; 4,5,7,8
The number of all possible samples of size 2 that can be drawn
from the population with replacement is N n = 42 = 16 , where
N, n are the population size and the sample size respectively.
The samples are:

(4, 4) (4, 5) (4, 7) (4, 8) (5, 4) (5, 5) (5, 7) (5, 8)

(7, 4) (7, 5) (7, 7) (7, 8) (8, 4) (8, 5) (8, 7) (8, 8)

The corresponding sample means X̄ are:

4.0 4.5 5.5 6.0 4.5 5.0 6.0 6.5

5.5 6.0 7.0 7.5 6.0 6.5 7.5 8.0

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 8 / 38
SAMPLING DISTRIBUTION OF MEANS
The frequency distribution of these mean values and their
probabilities is given as:

Possible values of X̄i Frequency Relative freq.(Probability)

4.0 1 0.0625
4.5 2 0.1250
5.0 1 0.0625
5.5 2 0.1250
6.0 4 0.2500
6.5 2 0.1250
7.0 1 0.0625
7.5 2 0.1250
8.0 1 0.0625
Total 16 1.0000
Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION
October 18, 2016 9 / 38
SAMPLING DISTRIBUTION OF MEANS

For the distribution of the sample means X̄i , the expected value
is
X
µX̄ =E (X̄ ) = x¯i p(x¯i )
=4.0(0.0625) + 4.5(0.1250) + 5.0(0.0625) + 5.5(0.1250)
+ 6.0(0.2500) + 6.5(0.1250) + 7.0(0.0625) + 7.5(0.1250)
+ 8.0(0.0625)
=6

which is identical with the population mean

(4 + 5 + 7 + 8)/4 = 6. Thus the mean of the sample means is
equal to the population mean.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 10 / 38
SAMPLING DISTRIBUTION OF MEANS
The variance of he distribution of the sample mean is
σX̄2 =E (X̄ )2 − (E (X̄ ))2
E (X̄ )2 =4.02 (0.0625) + 4.52 (0.1250) + 5.02 (0.0625)
+ 5.52 (0.1250) + 6.02 (0.2500) + 6.52 (0.1250)
+ 7.02 (0.0625) + 7.52 (0.1250) + 8.02 (0.0625)
=149/4

149
⇒ σX̄2 = E (X̄ )2 − (E (X̄ ))2 =
− 62 = 1.25
6
The standard deviation of the distribution of the means is
√
σX̄ = 1.25 = 1.118
Take note that this variance value equals the population variance
2
σ 2 = 2.5 divided by the sample size n = 2. Thus ( σn ).
Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION
October 18, 2016 11 / 38
SAMPLING DISTRIBUTION OF MEANS
THEOREM
Let X be a random variable with E (X ) = µ and variance
Var (X ) = σ 2 . Let X̄ be the sample mean of a random sample of size
n. Then the mean of the sampling distribution of means is given by

µX̄ = E (X̄ ) = µ

thus, the expected value of the sample mean is the population mean.
PROOF
X1 + X2 + ... + Xn
X̄ =
n
1
E (X̄ ) = [E (X1 ) + E (X2 ) + ... + E (Xn )]
n
1
= (nµ) = µ
n
Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION
October 18, 2016 12 / 38
SAMPLING DISTRIBUTION OF MEANS

THEOREM
If a population is infinite or if a sampling is with replacement, then
the variance of the sampling distribution of means, denoted by σX̄2 is
given by
σ2
Var (X̄ ) = σX̄2 =
n
thus, the variance of the sampling distribution is equal to the
population variance divided by the size of the sample used to obtain
the sampling distribution.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 13 / 38
SAMPLING DISTRIBUTION OF MEANS
PROOF
 
X1 X2 Xn
Var (X̄ ) = + + ... +
n n n
1
= [Var (X1 ) + Var (X2 ) + ... + Var (Xn )]
n2
1 σ2
= 2 (nσ 2 ) =
n n

STANDARD ERROR OF THE MEAN

The positive square root of the variance of the mean is referred to as
the standard error of the sample mean.
σ
σX̄ = √
n
Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION
October 18, 2016 14 / 38
SAMPLING DISTRIBUTION OF MEANS

EXAMPLE
Referring to the previous example with population data 4,5,7,8 where
different samples of sizes 2 each were drawn from the population,
find the means of X̄ and its standard error.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 15 / 38
SAMPLING DISTRIBUTION OF MEANS

SOLUTION
From the population µ = 6 and σ 2 = 2.5
E (X̄ ) = µ = 6
√
σ2 2.5 √2.5
Var (X̄ ) = σX̄2 = n
= 2
⇒ σX̄ = 2
= 1.118

NB
So far we have only shown that the sample mean X̄ has a
distribution, without necessarily specifying what form of a known
distribution it follows. However, by the central limit theorem, we
shall be able to say that the sampling distribution of the mean follow
a particular known distribution.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 16 / 38
Outline of Presentation

1 INTRODUCTION

2 SAMPLING DISTRIBUTION OF MEANS

3 THE CENTRAL LIMIT THEOREM

4 SAMPLING DISTRIBUTION OF PROPORTIONS

5 SAMPLING DISTRIBUTION OF THE VARIANCE

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 17 / 38
THE CENTRAL LIMIT THEOREM

Let X̄ denote the mean of a random sample of size n from a

population with mean µ and variance σ 2 . Then, X̄ is approximately
2
N(µ, σn ) when n is large. Thus, if

X̄ − µ
Z=
√σ
n

then, Z is approximately N(0,1) for large values of n(n ≥ 30).

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 18 / 38
THE CENTRAL LIMIT THEOREM

EXAMPLE
Suppose that the I.Q is a random sample with mean µ = 100 and
standard deviation σ = 25. What is the probability that in a class of
40 students the average I.Q. exceeds 110.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 19 / 38
THE CENTRAL LIMIT THEOREM

SOLUTION
25
n = 40, µ = 100, σ = 25, σX̄ = √ = 3.9528
40
By the central limit theorem, the mean I.Q. of 40 students is
normally distributed. Hence
110 − 100
P(X̄ ≥ 110) =P(Z ≥ )
3.9528
=P(Z ≥ 2.53)
=1 − φ(2.53)
=1 − 0.9943
=0.0057

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 20 / 38
THE CENTRAL LIMIT THEOREM

EXAMPLE
If the mean and standard deviation of serum iron values for healthy
men are 120 and 15 micrograms per 100 ml respectively, what is the
probability that a random sample of 50 normal men will yield a mean
between 115 and 125 micrograms per 100 ml?

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 21 / 38
THE CENTRAL LIMIT THEOREM

SOLUTION
15
n = 50, µ = 120, σ = 15, σX̄ = √
50
By the central limit theorem, the mean of the serum iron values for a
sample of 50 men is normally distributed. Hence
115 − 120 125 − 120
P(115 ≤ X̄ ≤ 125) =P( ≤Z ≤ )
√15 √15
50 50
=P(−2.36 ≤ Z ≤ 2.36)
=φ(2.36) − φ(−2.36)
=0.9909 − 0.0091
=0.9818

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 22 / 38
WHAT IF THE VARIANCE OF THE POPULATION IS NOT
KNOWN?

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 23 / 38
SAMPLING DISTRIBUTION OF THE MEAN(σ
unknown)

Application of the central limit theorem requires knowledge of

the population standard deviation σ.

If σ is not known, we can replace σ by S, the sample standard

deviation and the central limit theorem will still be valid.

However, when this is done, the distribution of the standardized

sample mean
X̄ − µ
√
S/ n
changes from the standard normal to the student’s t distribution.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 24 / 38
SAMPLING DISTRIBUTION OF THE MEAN(σ
unknown)

THEOREM
Let X̄ denote the mean of a random sample of size n from a normal
populationPwith mean µ and unknown variance σ 2 . Let
1 n
S 2 = n−1 2
i=1 (X − X̄ ) . Then,

X̄ − µ
T = √
S/ n

has a t-distribution wiht n-1 degrees of freedom.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 25 / 38
SAMPLING DISTRIBUTION OF THE MEAN(σ
unknown)

PROOF
n Pn 2
2 1 X 2 (n − 1)S 2 i=1 (X − X̄ )
S = (X − X̄ ) ⇒ =
n − 1 i=1 σ2 σ2
n  2 X n
X X − X̄ (n − 1)S 2
Z2 ⇒ ∼ χ2n−1


= = 2
i=1
σ i=1
σ

The degree of freedom is n − 1 because of the corrected S 2 where

the sample size is n − 1 instead of n. This is so because we computed
for one statistic, x̄ before finding S 2 .

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 26 / 38
SAMPLING DISTRIBUTION OF THE MEAN(σ
unknown)

PROOF CNT’D

X̄ − µ X̄ − µ X̄ − µ
T = =q =
√S
r
(n−1)S 2 σ2 (n−1)S 2
n × σ σ2
σ2 n(n−1) √ (n−1)
n
X̄ −µ
σ
√
n Z
=r =q ∼ tn−1
(n−1)S 2 χ2n−1
σ2 (n−1)
(n−1)

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 27 / 38
SAMPLING DISTRIBUTION OF THE MEAN(σ
unknown)

let us consider the previous example but this time around without a
known σ.
EXAMPLE
Suppose that the I.Q is a random sample with mean µ = 100. What
is the probability that in a class of 30 students with x̄ = 96 and
s = 21.6, the average I.Q. exceeds 110.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 28 / 38
SAMPLING DISTRIBUTION OF THE MEAN(σ
unknown)

SOLUTION
s 21.6
n = 40, µ = 110, s = 21.6, √ = √ = 3.4152
n 40

X̄ − 100
⇒ P(X̄ ≥ 100) =P(T ≥ )
√s
n
110 − 100
⇒ P(X̄ ≥ 100) =P(T ≥ )
3.4152
=P(T ≥ 2.53)
=0.00327178

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 29 / 38
Outline of Presentation

1 INTRODUCTION

2 SAMPLING DISTRIBUTION OF MEANS

3 THE CENTRAL LIMIT THEOREM

4 SAMPLING DISTRIBUTION OF PROPORTIONS

5 SAMPLING DISTRIBUTION OF THE VARIANCE

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 30 / 38
SAMPLING DISTRIBUTION OF PROPORTIONS

A sample proportion p̂ may be considered as a proportion of

successes which is obtained by dividing the number of successes
by the sample size n.

THOEREM
If sampling is with replacement, then
a E (p̂) = p

b Var (p̂) = p(1−p)

n
where p̂ is the sample proportion.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 31 / 38
SAMPLING DISTRIBUTION OF PROPORTIONS

THOEREM
The variable
p̂ − p
q
σp̂2
approaches the Standard normal distribution when n becomes large
such that np̂, n(1 − p̂) > 5.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 32 / 38
SAMPLING DISTRIBUTION OF PROPORTIONS

EXAMPLE
A consignment of bulbs has 20% defective. If a random sample of
250 is drawn with replacement from this consignment, what is the
probability that 22% will be defective?

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 33 / 38
SAMPLING DISTRIBUTION OF PROPORTIONS
SOLUTION

n = 250, p = 0.2, we are required to calculate P(p̂ ≤ 0.22)

 
p̂ − 0.2 0.22 − 0.2 
P(p̂ ≤ 0.02) =P  q ≤q
0.2(1−0.2) 0.2(1−0.2)
250 250
 
0.22 − 0.2 
P(p̂ ≤ 0.02) =P Z ≤ q
0.2(1−0.2)
250

=P(Z ≤ 0.7906)
=0.7854113

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 34 / 38
Outline of Presentation

1 INTRODUCTION

2 SAMPLING DISTRIBUTION OF MEANS

3 THE CENTRAL LIMIT THEOREM

4 SAMPLING DISTRIBUTION OF PROPORTIONS

5 SAMPLING DISTRIBUTION OF THE VARIANCE

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 35 / 38
SAMPLING DISTRIBUTION OF THE VARIANCE

If X1 , X2 , ..., Xn denote the random sample j of size n, then the

random variable sj2 which denotes the sample variance is defined
by

(X1j − X̄j )2 + (X2j − X̄j )2 + ... + (Xnj − X̄j )2

sj2 =
n−1
Pn 2
(X − X̄ )
= i=1
n−1
where Xnj for instance means the nth value of the jth sample
and X̄j is the mean of the jth sample.

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 36 / 38
SAMPLING DISTRIBUTION OF THE VRIANCE
With reference to the example for the sampling distribution of
the mean for instance, suppose a population consists of four
members: 4,5,7,8; the corresponding sample variances sj2 for
each of the 16 samples that can be drawn are:
0.00 0.50 4.50 8.00 4.50 0.00 2.00 4.50
4.50 2.00 0.00 0.50 8.00 4.50 0.50 0.00
a The mean of this sampling distribution of variances is
k
1 X 2
E (sj2 ) = µsj2 = s = 2.5
16 j=1 j
b The variance of this sampling distribution of variances is
Pk Pk
j=1 (sj2 − µsj2 )2 2
j=1 (sj − µsj2 )
2
2
Var (sj ) = = = 7.375
Nn 16
Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION
October 18, 2016 37 / 38
SAMPLING DISTRIBUTION OF THE VARIANCE

THEOREM
Pn
(X −X̄ )2
If S 2 = i=1n−1 is the variance of a random sample of size n taken
from a normal population with variance σ 2 then
Pn 2
(n − 1)S 2 i=1 (X − X̄ )
=
σ2 σ2
has a chi square distribution with v = n − 1 degrees of freedom.

PROOF
Refer to sampling distribution of mean(σ unknown)

Gabriel Asare Okyere (Ph.D.) () DISTRIBUTIONS RELATED TO THE NORMAL DISTRIBUTION

October 18, 2016 38 / 38