Special Continuous Probability
Distributions
Normal Distribution
Instructor: Dr. Gaurav Bhatnagar
22001: PROBABILITY AND STATISTICS
Normal Distribution - Definition
A normal distribution curve is symmetrical, bell-shaped
curve defined by the mean and standard deviation of a
data set.
The normal curve is a probability distribution with a total
area under the curve of 1.
The normal distribution often called Gaussian
Distribution or Bell Shaped Distribution.
Normal Distribution - Definition
A randomvariable is said to have a normal distribution
with mean and standard deviation , if the probability
density function is given by
Notation: or
X
o
2
1
2 1
( ) , (1)
2
x
f x e x
o
o t
| |
| |
|
|
|
\ .
\ .
= s s
2
~ ( , ) X N o
If and , the normal distribution reduces
to Standard Normal Distribution.
0 =
~ ( , ) X N o
2
1( 1) o o = =
Normal Distribution - Definition
Normal distributions are bell shaped, and symmetrical around .
Why symmetrical? Let = 100. Suppose x = 110.
2 2
10
) 2 / 1 (
100 110
) 2 / 1 (
e
2
1
e
2
1
) 110 ( f
|
.
|
\
|
o
|
.
|
\
|
o
t o
=
t o
=
Now suppose x = 90
2 2
10
) 2 / 1 (
100 90
) 2 / 1 (
e
2
1
e
2
1
) 90 ( f
|
.
|
\
|
o
|
.
|
\
|
o
t o
=
t o
=
110 90
Normal Distribution - Definition
The total area under every normal curve is 1.
The expected value (also called the mean) or can
be any number.
The standard deviation can be any nonnegative
number.
There are infinitely many normal distributions.
( ) E X
o
Normal Distribution - Definition
Total area is 1; symmetric around
1
2
1
2
(=0,
2
=1) and (=5,
2
=1)
-2 0 2 4 6 8
x
0
.
0
0
.
1
0
.
2
0
.
3
0
.
4
f
x
1
Normal Distribution - Definition
A B
-3 -2 -1 0 1 2 3
x
0
.
0
0
.
1
0
.
2
0
.
3
0
.
4
y
(=0,
2
=1)
Normal Distribution - Definition
(=0,
2
=2)
A
B
-2 0 2 4 6 8
x
0
.
0
0
.
2
0
.
4
0
.
6
0
.
8
1
.
0
f
x
1
(=0,
2
=1)
Normal Distribution - Definition
(=2,
2
=0.25)
A
B
How does the standard deviation affect the shape of f(x)?
o= 2
o =3
o =4
= 10 = 11 = 12
How does the expected value affect the location of f(x)?
Normal Distribution - Definition
The effect of and on pdf. o
No matter what and o are, the area between -o and
+o is about 68%; the area between -2o and +2o is
about 95%; and the area between -3o and +3o is
about 99.7%. Almost all values fall within 3 standard
deviations.
Normal Distribution - Definition
The beauty of Normal Curve
68% of
the data
95% of the data
99.7% of the data
Normal Distribution - Definition
68-95-99.7 Rule
-o +o +2o +3o -3o -2o
2
2
2
1
2
1
2
2
2
1
3
2
3
1
.68
2
1
.95
2
1
.997
2
x
x
x
e dx
e dx
e dx
o
o
o
o
o
o
o
o
o
o t
o t
o t
+ | |
|
\ .
+ | |
|
\ .
+ | |
|
\ .
=
=
=
}
}
}
Normal Distribution - Definition
68-95-99.7 Rule: Mathematical Perspective
Normal Distribution - Definition
Howgood this rule is for real data?
Let us take an example of the marks in Mid-Sem1 of the
Probability and Statistics class.
Observations are: Total Students =146
Mean weight =9.22
Standard deviation =4.36
According to this rule:
o 68% of 146 = 0.68x146 = 99.28 ~ 99 Students
o 95% of 146 = 0.95 x 146 = 138.7 ~ 138 Students
o 99.7% of 146 = 0.997 x 146 = 145.5 ~ 145 Students
Real Facts:
o Rule says: 68% of 146 = 99 Students. In fact, 96
students fall within (-o, +o).
o Rule says: 95% of 146 = 138 Students. In fact, 136
students fall within (-2o, +2o).
o Rule says: 99.7% of 146 = 145 Students. In fact, 145
students fall within (-3o, +3o).
Normal Distribution - Definition
Normal Distribution - Definition
Moments:
Mean: r = 1
Variance:
2
1
2 1
' ( ) (2)
2
x
r r
r
x f x dx x e dx
o t
| |
| |
|
|
|
\ .
\ .
= =
} }
2
1
2
1
1
' (3)
2
x
xe dx
o
o t
| |
| |
|
|
|
\ .
\ .
= =
}
( )
( ) ( )
2 2
2 2
2 1
( ) ( ) ' ' (4) V X E X E X o = = =
Normal Distribution - Definition
Moment Generating Function:
( )
2
2 2
1
2
1
2
( )
1
2
(5)
xt
X
x
xt
t t
M t E e
e e dx
e
o
o
o t
| |
| |
|
|
|
\ .
\ .
| |
+
|
\ .
=
=
=
}
