Normal (Gaussian) Distribution

Continuous Data Sampling Distribution (STU 07215)

Leguma Bakari
Email: leguma.bakari@eastc.ac.tz
Email: leguma2020@gmail.com
Phone:+255 762 760 095
September 3, 2022
Eastern Africa Statistical Training Center (EASTC)

1
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

2
Definition

• Normal distribution, also known as the Gaussian distribution, is a

probability distribution that is symmetric about the mean.
• The distribution show that, data near the mean are more frequent
in occurrence than data far from the mean.
• The normal distribution is the most important probability
distribution in statistics because it fits many natural phenomena.
• For example, heights, blood pressure, measurement error, and IQ
scores follow the normal distribution.
• A random variable X is said to follow Normal distribution with
parameters 𝜇 and 𝜎 2 if its density function satisfy the following
function
•
− 1 ( x − 𝜇) 2
√1 e 2 𝜎 2



 ; −∞ < x < ∞
f ( x; 𝜇, 𝜎 2 ) = 𝜎 2 𝜋
 0, otherwise
 3

• The above density function can be indicated as X ∼ N (𝜇, 𝜎 2 )
• In graph form, normal distribution will appear as a bell curve.

Figure 1: Normal Curve

4
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

5
Parameters of the Normal Distribution

• As with any probability distribution, the parameters for the normal distribution
define its shape and probabilities entirely.
• The normal distribution has two parameters, the mean and standard deviation
or variance.
• The normal distribution does not have just one form.
• Instead, the shape changes based on the parameter values, as showing below.

(a) Different Means-Same (b) Same Means-different

Standard Deviations Standard Deviations
6
Figure 2: Mean and SD Displays
Mean and Standard Deviation

Mean (Location Parameter)

• The mean is the central tendency of the distribution.

• It defines the location of the peak for normal distributions.
• Most values cluster around the mean.
• On a graph, changing the mean shifts the entire curve left or right on the
horizontal-axis.

Standard deviation (Scale Parameter)

• The standard deviation is a measure of variability.

• It defines the width of the normal distribution.
• The standard deviation determines how far away from the mean the values tend
to fall.
• It represents the typical distance between the observations and the average.
• On a graph, changing the standard deviation either tightens or spreads out the
width of the distribution along the horizontal-axis.
7
• Larger standard deviations produce distributions that are more
spread out and vice versa.
• When you have narrow distributions, the probabilities are higher
that values won’t fall far from the mean.
• As you increase the spread of the distribution, the likelihood that
observations will be further away from the mean also increases.

8
Figure 3: Normal Curves 9
Function of the Parameters

• Mean is the location parameter.

• The standard deviation is the scale parameter.

10
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

11
Symmetric Property for All Forms of Normal Distribution

• As already outlined normal curve is always symmetric.

• That is the mean, median, and mode are all equal.
• The normal distribution cannot model skewed distributions.
• The symmetrical property allow the application of empirical rule.
• The Empirical Rule allows you to determine the proportion of
values that fall within certain distances from the mean.

12
Figure 4: Symmetric Distribution

13
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

14
The Proof of Moment Generating Function

• Mx ( t ) = E ( etx )
∫∞
• Mx ( t ) = E ( etx ) = −∞ etx f ( x; 𝜇, 𝜎 2 ) dx
• Recall;
∫∞ − 1 ( x − 𝜇) 2
f ( x; 𝜇, 𝜎 2 ) = −∞ etx √1 e 2 𝜎2 , for − ∞ < x < ∞
𝜎 2𝜋
1
∫∞ − ( x − 𝜇) 2
• Therefore; Mx ( t ) = −∞ etx √1 e 2 𝜎2 dx
𝜎 2𝜋
∞ 1 2
− ( x − 𝜇)
• Mx ( t ) = √1
∫
etx e 2 𝜎2 dx
𝜎 2 𝜋 −∞
1
∞ [ 2 𝜎 tx − ( x − 𝜇) 2 ]
2
• Mx ( t ) = √1
∫
−∞
e 2 𝜎2 dx
𝜎 2𝜋
1
∞ [ 2 𝜎 tx − ( x − 2 𝜇 x +𝜇 2 ) ]
2 2
• Mx ( t ) = √1
∫
e 2 𝜎2 dx
𝜎 2 𝜋 −∞
∞ 1 2 2 2
[ 2 tx − x + 2 x − ]
• Mx ( t ) = √1
∫ 𝜎 𝜇 𝜇
e 2 𝜎2 dx
𝜎 2 𝜋 −∞
∞ 1 2 2 2
[ − x + 2 tx + 2 x − ]
• Mx ( t ) = √1
∫ 𝜎 𝜇 𝜇
e 2 𝜎2 dx
𝜎 2 𝜋 −∞
∞ 1 2 2 2
− [ x − 2 𝜎 tx − 2 𝜇 x +𝜇 ]
• Mx ( t ) = √1
∫
−∞
e 2 𝜎2 dx
𝜎 2𝜋
15
 − 1 [ x 2 − 2 ( 𝜎 2 t +𝜇) x +𝜇 2 ]
∫∞
• Mx ( t ) = √1
−∞
e 2 𝜎2 dx
𝜎 2𝜋
1
∞ − [ x − 2 ( 𝜎 t +𝜇) x +( 𝜎 t +𝜇) 2 − ( 𝜎 2 t +𝜇) 2 +𝜇 2 ]
2 2 2
• Mx ( t ) =
∫
√1 e 2 𝜎2 dx
𝜎 2𝜋 −∞
• Mx ( t ) =
1
1
∫∞ − { [ x 2 − 2 ( 𝜎 2 t +𝜇) x +( 𝜎 2 t +𝜇) 2 ] − [ ( 𝜎 2 t +𝜇) 2 − 𝜇 2 ] }
√
−∞
e 2 𝜎2 dx
𝜎 2𝜋
• consider
• (𝜎 2 t + 𝜇) 2 − 𝜇2 = (𝜎 2 t ) 2 + 2 𝜇𝜎 2 t + 𝜇2 − 𝜇2 = (𝜎 4 t 2 + 2 𝜇𝜎 2 t )
and
• [x 2 − 2 (𝜎 2 t + 𝜇) x + (𝜎 2 t + 𝜇) 2 ] = [x − (𝜎 2 t + 𝜇)] 2 , therefore
∫ ∞ − 1 { [ x − ( 𝜎 2 t +𝜇) ] 2 − ( 𝜎 4 t 2 +2 𝜇 𝜎 2 t ) }
• Mx ( t ) = √1 e 2 𝜎2 dx
𝜎 2 𝜋 −∞
1 2 2 1
∞ { − 2 [ x − ( 𝜎 t +𝜇) ] + 2 ( 𝜎 t +2 𝜇 𝜎 t ) } 4 2 2
• Mx ( t ) = √1
∫
e 2𝜎 2𝜎 dx
𝜎 2 𝜋 −∞
∞ 1 2 2 1 4 2 2
− [ x − ( t +𝜇) ] ( t + 2 t )
• Mx ( t ) = √1
∫ 𝜎 𝜎 𝜇 𝜎
e 2 𝜎2 e 2 𝜎2 dx
𝜎 2 𝜋 −∞
1 4 2 2 ∞ 1 2 2
( t + 2 t ) − [ x − ( t +𝜇) ]
• Mx ( t ) = √1 e 2 𝜎2
𝜎 𝜇 𝜎 ∫ 𝜎
−∞
e 2 𝜎2 dx
𝜎 2𝜋
1 4 2 2 ∞ 1 2 2
( 𝜎 t +2 𝜇 𝜎 t ) − [ x − ( 𝜎 t +𝜇) ]
• Mx ( t ) = e 2 𝜎2
∫
√1 e 2 𝜎 2 dx
−∞ 𝜎 2𝜋
• which is X ∼ N (𝜇 + 𝜎 2 t , 𝜎 2 ), the area under the curve is
16
 1
∫∞ − [ x − ( 𝜎 2 t +𝜇) ] 2
• √1
−∞ 𝜎 2 𝜋
e 2 𝜎2 dx = 1, therefore
1 𝜎 4 t 2 +2 𝜇 𝜎 2 t 2 𝜎 2 ( 1 𝜎 2 t 2 +𝜇 t )
( 𝜎 4 t 2 +2 𝜇 𝜎 2 t ) 2
• Mx ( t ) = e 2 𝜎2 ( 1) = e 2 𝜎2 =e 2 𝜎2

2 𝜎 2 ( 1 𝜎 2 t 2 +𝜇 t )
2 1 2 2
• Mx ( t ) = e 2 𝜎2 = e2 𝜎 t +𝜇 t

• The moment generating function for Normal distribution is given

by
1 2 2
• Mx ( t ) = e 𝜇t + 2 𝜎 t

17
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

18
Expected Value

• Recall; E ( X ) = Mx′ ( t )| t =0
1 2 2
• Mx ( t ) = e 𝜇t + 2 𝜎 t

1 2 2
• Mx′ ( t ) = (𝜇 + 𝜎 2 t ) e 𝜇t + 2 𝜎 t

1 2
(0) 2
• E ( X ) = Mx′ ( t )| t =0 = (𝜇 + 𝜎 2 ( 0)) e 𝜇 ( 0 )+ 2 𝜎
• E ( X ) = (𝜇 + 0) e ( 0+0 ) = 𝜇e0 = 𝜇( 1) = 𝜇
• The expected value or mean for Normal distribution is given by
E (X ) = 𝜇

19
Variance

• Var ( X ) = E ( X 2 ) − [ E ( X )] 2
• E ( X 2 ) = Mx′′ ( t )| t =0
1 2 2
• Mx′ ( t ) = (𝜇 + 𝜎 2 t ) e 𝜇t + 2 𝜎 t

• By product rule method of differentiation

1 2 2 1 2 2
• Mx′′ ( t ) = (𝜇 + 𝜎 2 t ) (𝜇 + 𝜎 2 t ) e 𝜇t + 2 𝜎 t + ( 0 + 𝜎 2 ) e 𝜇t + 2 𝜎 t
1 2 2 1 2 2
• Mx′′ ( t ) = (𝜇 + 𝜎 2 t ) 2 e 𝜇t + 2 𝜎 t + 𝜎 2 e 𝜇t + 2 𝜎 t
• E ( X 2 ) = Mx′′ ( t )| t =0 =
1 2 2 1 2 2
(𝜇 + 𝜎 2 ( 0)) 2 e 𝜇 ( 0 )+ 2 𝜎 ( 0 ) + 𝜎 2 e 𝜇 ( 0 )+ 2 𝜎 ( 0 )
• E ( X 2 ) = (𝜇 + 0) 2 e0 + 𝜎 2 e0 = 𝜇2 ( 1) + 𝜎 2 ( 1)
• E ( X 2 ) = 𝜇2 + 𝜎 2
• Var ( X ) = E ( X 2 ) − [ E ( X )] 2
• Var ( X ) = (𝜇2 + 𝜎 2 ) − (𝜇) 2 = 𝜇2 + 𝜎 2 − 𝜇2
• Var ( X ) = 𝜎 2
• The variance for Normal distribution is given by Var ( X ) = 𝜎 2 20
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

21
The Standard Normal Distribution

• Normal distributions do not necessarily have the same means and

standard deviations.
• The standard normal distribution is a special case of the normal
distribution.
• It is the distribution that occurs when a normal random variable
has a mean of zero and a standard deviation of one.
• The normal random variable of a standard normal distribution is
called a standard score or a Z score.
• Every normal random variable X can be transformed into a Z
score via the following equation:
X −𝜇
• Z = 𝜎 where
• X is a normal random variable,
• 𝜇 is the mean, and
• 𝜎 is the standard deviation.
22
Probability Density Function for Standard Normal Distribution

• Recall a Normal distribution density function

•
− 1 ( x − 𝜇) 2
√1 e 2 𝜎 2



 ; −∞ < x < ∞
f ( x; 𝜇, 𝜎 2 ) = 𝜎 2 𝜋
 0, otherwise


• which can be re-written as
•
1 x −𝜇 2
 √1 e− 2 ( 𝜎 ) ; −∞ < x < ∞


2

f ( x; 𝜇, 𝜎 ) = 𝜎 2 𝜋
 0, otherwise


−𝜇
• For Standard Normal distribution Z = X 𝜎 and 𝜎 = 1, therefore
the density function will be given by
•
1 2
 √1 e− 2 z ; −∞ < z < ∞



f (z) = 2 𝜋
 0, otherwise

23

Moment Generating Function for Standard Normal Distribution

• Recall the moment generating function for Normal distribution

which is given by
1 2 2
• Mx ( t ) = e 𝜇t + 2 𝜎 t

• For standard normal distribution, 𝜇 = 0 and 𝜎 2 = 1, therefore

1 2 2 1 2 1 2 1 2
• Mx ( t ) = e 𝜇t + 2 𝜎 t
= e ( 0 ) t + 2 ( 1 ) t = e0+ 2 t = e 2 t
• The moment generating function for Standard Normal
1 2
distribution is given by Mx ( t ) = e 2 t

24
Mean of the Standard Normal Distribution

• The mean of Standard Normal Distribution random variable is

always zero, proof,
−𝜇
• Recall; Z = X 𝜎
 
−𝜇 E (X ) − 𝜇
• E (Z ) = E X 𝜎 = 𝜎 , but E ( X ) = 𝜇, therefore
E ( X ) −𝜇
• E (Z ) = 𝜎 = 𝜇− 𝜇
𝜎 = 0
𝜎 =0
• The mean of Standard Normal Distribution random variable is
given by E ( Z ) = 0

25
Variance of the Standard Normal Distribution

• The variance of Standard Normal Distribution random variable is

always one (unit), proof,
−𝜇
• Recall that; Z = X 𝜎
   
−𝜇
• Var ( Z ) = Var X 𝜎 = Var X𝜎 = 1
𝜎2
Var ( X )

• but Var ( X ) = 𝜎 2 ,therefore

𝜎2
• Var ( Z ) = 1
𝜎2
Var ( X ) = 1
𝜎2
(𝜎 2 ) = 𝜎2
=1
• The variance of Standard Normal Distribution random variable is
given by Var ( Z ) = 1

26
Standard Normal Cumulative Distribution Function

• Suppose X is a normal random variable with mean 𝜇 and

variance 𝜎 2 , then the distribution function is given by
−𝜇
• F (x ) = P (X < x ) = P ( X 𝜎 < x −𝜎𝜇 ) = P ( Z < z )
x−𝜇
• Φ( z ) = Φ( 𝜎 )
• z = x −𝜎𝜇 is the z-value obtained by standardizing x.
• ∫ z
1 1 2
Φ( z ) = P ( Z < z ) = √ e − 2 u du
2𝜋 −∞

27
Figure 6: Cumulative Probabilities
Types of Normal Curves (Area Under the Curve)

28
Figure 7: Area Under the Curve
Practical Examples

Recall EXERCISES FOR SECTION 4-6,page 117 download this

book from here.

29
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

30
The Proof of Equality of Mean, Median, and Mode

• Symmetric is the important property of Normal distribution is

which is simply the equality of three measure of central tendency
which are mean, mode and median.

Mean

• The mean of Normal distribution is given by the expected value

of the Normal distribution .
• The expected value for Normal distribution has already proved by
using moment generating function and it is expressed as
• E (X ) = 𝜇

31
Mode of Normal Distribution

• For Normal distribution ,mode is the value of random variable x

for which the density function f ( x ) is at maximum.
• There are two conditions for a random variable x to be a
maximum value for f ( x ).
1 The necessary condition is f ′ ( x ) = 0 and
2 The sufficient condition is f ′′ ( x ) < 0
• Recall the Normal distribution density function is given by
1
− ( x − 𝜇) 2
• f ( x ) = √1 e 2 𝜎2
𝜎 2𝜋
• Apply naturalnlogarithm to the density function
1
( x − 𝜇) 2
o
−
• ln [f ( x )] = ln √1 e 2 𝜎2
𝜎 2𝜋 
[ − 1 ( x − 𝜇) 2 ]

• ln [f ( x )] = ln √1 + ln e 2 𝜎2
  𝜎 2𝜋
• let √1 = c where c is a constant.
𝜎 2𝜋
• ln [f ( x )] = c − 1
2 𝜎2
(x − 𝜇) 2 ln e = c − 1
2 𝜎2
(x − 𝜇) 2
32
• apply differentiation with respect to x
d ln [ f ( x ) ]
• dx
2
= 0 − 2𝜎 2 (x − 𝜇) ( 1)
′
f (x )
• f (x )
= − 𝜎12 ( x − 𝜇)
• f ( x ) = − 𝜎 2 ( x − 𝜇) f ( x )
′ 1

• for maximum or minimum f ′ ( x ) = 0

• − 𝜎12 ( x − 𝜇) f ( x ) = 0
• For the above expression to become zero, then either ( x − 𝜇) or
f ( x ) should be zero or both,
• but f ( x ) ≠ 0, i.e f ( x ) can never become zero since x is the
exponent, therefore
• for f ′ ( x ) to become zero the expression ( x − 𝜇) must be equal to
zero.
• − 𝜎12 ( x − 𝜇) = 0, ( x − 𝜇) = 0, therefore x = 𝜇.
• for maximum f ′ ( x ) < 0, recall
33
• f ′ ( x ) = − 𝜎12 ( x − 𝜇) f ( x )
• by using product rule
• f ′′ ( x ) = − 𝜎12 [( x − 𝜇) f ′ ( x ) + ( 1) f ( x )], but
• f ′ ( x ) = − 𝜎12 ( x − 𝜇) f ( x ), therefore
h   i
• f ′′ ( x ) = − 𝜎12 ( x − 𝜇) − 𝜎12 ( x − 𝜇) f ( x ) + f ( x )
h i
• f ′′ ( x ) = − 𝜎12 − 𝜎12 ( x − 𝜇) 2 f ( x ) + f ( x )
h i
f (x )
• f ′′ ( x ) = − 𝜎 2 − 𝜎12 ( x − 𝜇) 2 + 1 , at x = 𝜇
h i h i
f (x ) f (x )
• f ′′ ( x ) = − 𝜎 2 − 𝜎12 (𝜇 − 𝜇) 2 + 1 = − 𝜎 2 − 𝜎12 ( 0) 2 + 1
• f ′′ ( x ) = − f 𝜎( x2) ( 0 + 1) = − f 𝜎( x2)
• f ′′ ( x ) = − f 𝜎( x2) , recall
1
− ( x − 𝜇) 2
• f (x ) = √1 e 2 𝜎2 , at x = 𝜇
𝜎 2𝜋
1 2 1
− ( 𝜇− 𝜇) − (0) 2
• f (x ) = √1 e 2 𝜎2 = √1 e 2 𝜎2 = √1 e0 = √1
𝜎 2𝜋 𝜎 2𝜋 𝜎 2𝜋 𝜎 2𝜋
√1
• Therefore, f ′′ ( x ) = − f 𝜎( x2) = − 𝜎𝜎22𝜋 = − 1
√
𝜎3 2 𝜋
34
• since 𝜎 can never become negative , then
• f ′′ ( x ) = − 1
√ <0
𝜎3 2 𝜋
• Therefore the mode of Normal distribution is 𝜇.

35
Median of Normal Distribution

• The density function for a Normal distribution is given by

1
− ( x − 𝜇) 2
• f ( x; 𝜇, 𝜎) = √1 e 2 𝜎2 ; −∞ < x < ∞
𝜎 2𝜋
• the area under the curve property is
∫∞ ∫∞ − 1 ( x − 𝜇) 2
• −∞ f ( x; 𝜇, 𝜎) dx = −∞ √1 e 2 𝜎2 dx = 1
𝜎 2𝜋
• Let M be median of a Normal distribution then
∫M − 1 ( x − 𝜇) 2 ∫∞ − 1 ( x − 𝜇) 2
• −∞ √1 e 2 𝜎2 dx = M √1 e 2 𝜎2 dx = 1
2
𝜎 2𝜋 𝜎 2𝜋
∫∞ 1 1 2
− ( x − 𝜇)
• Consider; M √ e 2 𝜎2 dx = 21
𝜎 2𝜋
∫∞ 1 x−𝜇 2
• M √1 e− 2 ( 𝜎 ) dx = 12
∫ ∞ 𝜎 2 𝜋 1 x−𝜇 2
• M √1 e− 2 ( 𝜎 ) dx =
∫ 𝜇 𝜎 12 𝜋 1 x−𝜇 2 ∫∞ 1 1 x −𝜇 2
√ e − 2 ( 𝜎 ) dx + √ e − 2 ( 𝜎 ) dx
M
∫ 𝜇 𝜎 12 𝜋 1 x − 𝜇 2 ∫ ∞ 𝜎 12 𝜋 1 x −𝜇 2
𝜇

• M √ e − 2 ( 𝜎 ) dx + 𝜇 √ e − 2 ( 𝜎 ) dx = 1
𝜎 2𝜋 𝜎 2𝜋 2

36
 ∫∞ 1x−𝜇 2
• consider √1 e − 2 ( 𝜎 ) dx
𝜇 𝜎 2𝜋
x−𝜇
• Let z = dz 1
𝜎 , then dx = 𝜎 , therefore dx = 𝜎 dz
• change of limits,
• for x = 𝜇, z = 𝜇−𝜎𝜇 = 0
• and x = ∞, z = ∞−𝜇𝜎 =∞
∫∞ 1 x−𝜇 2 ∫∞ 1 1 2 ∫∞ 1 1 2
• 𝜇 √1 e− 2 ( 𝜎 ) dx = 0 √ e − 2 z dz =
0
√ e − 2 z dz
𝜎 2𝜋 𝜎 2𝜋 2𝜋
• recall standard normal density function
∫∞ 1 2
• 0 √1 e− 2 z dz = 12
2𝜋
∫∞ 1 x−𝜇 2 ∫∞ 1 2
• 𝜇 √1 e− 2 ( 𝜎 ) dx = 0 √1 e− 2 z dz = 12
𝜎 2𝜋 2𝜋
1 x−𝜇 2 ∫∞ 1 x−𝜇 2
• Recall, M √1 e− 2 ( 𝜎 ) dx + 𝜇 √1 e− 2 ( 𝜎 ) dx = 21
∫𝜇
𝜎 2𝜋 𝜎 2𝜋
1 x−𝜇 2
• M √1 e− 2 ( 𝜎 ) dx + 12 = 12
∫𝜇
∫ 𝜇 𝜎 2 𝜋 1 x−𝜇 2
• M √1 e− 2 ( 𝜎 ) dx = 12 − 21 = 0
∫ 𝜇 𝜎 2 𝜋 1 x−𝜇 2
• M √1 e− 2 ( 𝜎 ) dx = 0
𝜎 2𝜋
• M = 𝜇, the median of Normal distribution is 𝜇
• For Normal distribution , mean = median = mode = 𝜇, proved. 37
Outline

1 Definition

2 Parameters of the Normal Distribution

3 Symmetric Property for All Forms of Normal Distribution

4 The Proof of Moment Generating Function

5 Expected Value and Variance

6 The Standard Normal Distribution

7 The Proof of Equality of Mean, Median, and Mode

8 Empirical Rule of the Normal Distribution

38
Empirical Rule for the Normal Distribution

• When you have normally distributed data, the mean and standard deviation
becomes particularly valuable.
• You can use them to determine the proportion of the values that fall within a
specified number of standard deviations from the mean.
• For example, in a normal distribution,
• 68.3% of the observations fall within ±1 sd from the mean,
• 95.4% of the observations fall within ±2 sd from the mean, and
• 99.7% of the observations fall within ±3 sd from the mean.
• This property is part of the Empirical Rule, which describes the percentage of
the data that fall within specific numbers of standard deviations from the mean
for bell shaped curves.
• The above explanation can be expressed mathematically as

P (𝜇 − 𝜎 < X < 𝜇 + 𝜎) = 0.6826

P (𝜇 − 2𝜎 < X < 𝜇 + 2𝜎) = 0.9544
P (𝜇 − 3𝜎 < X < 𝜇 + 3𝜎) = 0.9973 39
(a) General (b) Specific

Figure 8: Empirical Rule Display

40
Empirical Rule Proof by Using Standard Normal

 
𝜇−𝜎−𝜇 X −𝜇 𝜇+𝜎−𝜇
P (𝜇 − 𝜎 < X < 𝜇 + 𝜎) = P < < = 0.6826
𝜎 𝜎 𝜎
 
𝜇−𝜎−𝜇 X −𝜇 𝜇+𝜎−𝜇  −𝜎 𝜎
P < < =P <Z < = 0.6826
𝜎 𝜎 𝜎 𝜎 𝜎
−𝜎 𝜎
P( < Z < ) = P (−1 < Z < 1) = 0.6826
𝜎 𝜎
Therefore

P (𝜇 − 𝜎 < X < 𝜇 + 𝜎) = P (−1 < Z < 1) = 0.6826

P (𝜇 − 2𝜎 < X < 𝜇 + 2𝜎) = P (−2 < Z < 2) = 0.9544
P (𝜇 − 3𝜎 < X < 𝜇 + 3𝜎) = P (−3 < Z < 3) = 0.9973

41
Figure 9: Empirical Rule

42