From Equation (3.2.

3), we see that the normal distribution has the cumulative

distribution function
1 w−μ 2
x

Pr( X ≤x )=F ( x)=

1 − ( ) dw
2 σ
∫e
√ 2 πσ 2 −∞ , ……………………………………………(3.2.4)
From this we obtain
1 w−μ 2
− ( b

Pr (a≤X ≤b )=F(b )−F (a )=

1
e
2 σ ) dw
∫
√ 2 πσ 2 a , ………………………………(3.2.5)

The integral in (3.2.4) cannot be evaluated by elementary methods, but can be

represented in terms of the integral

z t2
1 −
Φ( z )= ∫ e 2 dt
√2 π −∞ ……………………………………………………………………(3.2.6)

which is the distribution function of the normal distribution with mean = 0 and
variance = 1 and has been tabulated (see mathematical tables).
Infact, if we set

w−μ dt 1
t= = ⇒ dw = σ dt
σ , then dw σ , and we have to integrate from
x−μ
−∞ to z=
σ .

From Eq. (3.2.4) we thus obtain

x− μ
σ t2
1 −
2
F( x)= ∫e σ dt
√2 πσ 2 −∞
x−μ
z=
σ drops out, and the expression on the right equal (3.2.6), where σ . That is

F( x )=Φ ( x−μσ ) ………………………………………………………………………..(3.2.7)

The cumulative distribution of z is known as the standard normal distribution.

From this important formula and Eq. (3.2.5) we obtain another important formula

Pr(a≤X ≤b )=F(b )−F (a )=Φ ( b−μσ ) − Φ( a−μσ )

Page 1 of 8
  Let X be binomial with parameters n and p. For large n, X is approximately
2
normal with mean, μ = np and variance σ = npq=np(1− p) .
For practical purposes, use normal approximation to binomial distribution for
either p > 0.5 and np > 5 or p > 0.5 and n (1-p) > 5.
 Let X be Poisson with parameter μ . Then for large values of n, X is
approximately normal with mean μ and variance μ .

Standard Normal Curve

Since a normal distribution could have an infinite number of values for its mean and
standard deviation, it is impossible to tabulate the area associated with each and every
normal curve. Instead, only a single curve is tabulated called the standard normal
distribution. The variate has a normal distribution with mean of zero and a standard
deviation of 1. The probability density function function of z is given by;
1 2
1 −2 z
f (z )= e
√2 π
Standard Normal Variable, Z

If a random variable X has a normal distribution with mean μ and standard

deviation σ then:
X−μ
Z=
σ
has a standard normal distribution. Z is known as a standardized normal variable.

X−μ
Z=
Let σ
We say
−z 2
1 2
f (z ) = e
√2 π

X −μ E( X )−E( μ)
E( Z ) = E
[ ]
σ
=
σ

But E( X )=μ

μ−μ
∴ E(Z ) = =0
σ

Page 2 of 8
 2 2
Var ( Z ) = E ( Z )−[ E( Z ) ]
= E( Z2 ) Since E( Z )=0

But

X−μ X −μ
E [ Z 2 ] =E [( ) ( )]
σ
⋅
σ

1 1
= 2
E [( X−μ )2 ] = 2 [ E ( X 2 −2 Xμ+ μ2 )]
σ σ

1
= 2
[ E( X 2 )−E (2 Xμ)+ E( μ2 )]
σ

1 2 2 1
= 2 [ E( X )−2 μE( X )+ μ ] = 2 [
E( X 2 )−2 μ2 + μ2 ]
σ σ

1 2 2 1 2 2
=
σ2
[ E( X )−μ ] =
σ2
[ E( X )− [ E ( X ) ] ]
1
= 2
⋅σ 2 =1
σ

∴ Var ( Z ) = 1

⇒ Z is normally distributed with mean = 0 and variance = 1.

i. e . Z ~ N (0, 1)

Example 3.2.3
Determine the probabilities
(a) Pr ( X ≤2. 44 ) , (b) Pr ( X ≤−1 .16 ) , (c) Pr ( X ≥1)
(d) Pr (2≤ X≤10 ) where X is assumed to be normal with mean 0 and variance 1.

Solution
2
Since μ=0 and σ =1 , we may obtain the desired values directly from the
mathematical tables;

Page 3 of 8
(a) 0.9927, (b) 0.1230, (c) Pr ( X ≥1)=1−Pr ( X< 1)=1−0 .8413=0 . 1587
(d) Pr ( 2≤ X≤10 )=Φ ( 10)−Φ(2)=1−0 . 9772=0 . 0228

Example 3.2.4
Let X be normal with mean 0 and variance 1. Determine the constant c such that:
(a) Pr ( X ≥c )=10 % , (b) Pr ( X ≤c )=5 % , (c) Pr (0≤X ≤c )=45 %
(d) Pr (−c≤ X≤c )=99 %

Solution

Since μ=0 and σ 2=1 , we have;

(a)

Pr ( X ≥c)=1−Pr ( X≤c )=1−Φ (c−0

1 )
=0 .1

⇒Φ (c)=1−0 .1=0 . 9 ⇒ c=Φ−1 (0 . 9)=1. 282

(b)
c−0
Pr ( X ≤c )=Φ ( )
1
=0. 05

⇒Φ(c )=0. 05 ⇒ c=Φ−1 (0 .05 )= −Φ−1 (1−0 . 05 )=−Φ−1 (0 . 95)=−1. 645

(c)
c−0 0−0
Pr (0≤X ≤c )=Φ ( ) ( )
1
−Φ
1
=Φ(c )−Φ(0 )=0 . 45

⇒Φ(c ) − 0 . 5=0 . 45 ⇒ Φ( c ) = 0 .5+0 . 45=0 . 95 ⇒c =Φ−1 (0. 95 )= 1. 645

Page 4 of 8
(d)

Pr (−c≤ X≤c )=Φ (c−0

1 ) − Φ(
−c−0
1 ) =Φ(c )−Φ(−c )=0. 99
⇒Φ(c) − [ 1−Φ(c )] =0.99 ⇒ 2Φ(c) = 1+0 .99=1.99 ⇒ Φ(c ) = 0.995

⇒ c=Φ−1 (0.995 )= 2 .576

Example 3.2.5
Let X be normal with mean = -2 and variance = 0.25. Determine the constant c such that:
(a) Pr ( X ≥c )=0 .2 , (b) Pr (−c≤ X≤−1 )=0 .5 ,
(c) Pr (−2−c≤X≤−2+ c )=0. 9 (d) Pr (−2−c≤X≤−2+ c )=99. 6 %

Solution

Since μ=−2 and σ 2=0.25 ⇒ σ=0.5 , we have;

(a)

Pr ( X ≥c )=1−Pr( X≤c )=1−Φ (c0+2.5 )=0 . 2

⇒Φ (c0+2. 5 )=1−0 .2=0 .8 ⇒
c+2
0 .5
=Φ−1 (0 . 8)=0 .842 ⇒ c=−1 .579

(b)

Pr (−c≤ X≤−1 )=Φ (−1+2

0 .5 ) − Φ(
−c+2
0 .5 )
=0 . 5

−c +2 −c +2 −c+2
⇒Φ (2) − Φ (0 .5 )
=0 . 5 ⇒ 0 . 9772−Φ
0.5 (
= 0.5 ⇒ Φ)0.5 (
=0.9772−0 .5 )
⇒Φ (−c+2
0.5 )=0. 4772 ⇒
−c +2
0 .5
=Φ −1
( 0. 4772 )= −Φ−1 (1−0 . 4772)=−Φ−1 (0. 5228 )

−c +2
⇒ = −0 . 057 ⇒ c=2 . 0285
0.5

Page 5 of 8
(c)

Pr (−2−c≤X ≤−2+c)=Φ (−2+c+2

0.5 ) − Φ(
−2−c+2
0.5 ) =0.9
⇒Φ(2c) − Φ (−2c )=0.9 ⇒ Φ(2c ) − [1−Φ ( 2c ) ]=0. 9 ⇒ 2Φ(2c) =1+0.9=1.9

⇒Φ ( 2c )=0.95 ⇒ 2c=Φ−1 (0.95)= 1 .645

⇒ c=0.823

(d)

Pr (−2−c≤X ≤−2+c)=Φ (−2+c+2

0.5 ) − Φ(
−2−c+2
0.5 ) =0.996
⇒Φ(2c) − Φ (−2 c )=0 .996 ⇒ Φ(2 c) − [ 1−Φ ( 2c ) ]=0 .996 ⇒ 2 Φ(2 c) =1+0.996=1.996

⇒ Φ ( 2c )=0 .998 ⇒ 2c=Φ−1 (0.998 )= 2 .878

⇒ c=1.439
Example 3.2.6
An industrial process mass produces items which are normally distributed. 11.55% of
them weigh over 20 kg and 5.89% weigh under 10 kg. Calculate the mean weight and
standard deviation for this distribution.

Solution

(a)
20−μ
Pr (X>20)=1−Pr( X≤20)=1−Φ ( )
σ
=0.1155

20−μ 20−μ −1
⇒Φ ( )
σ
=1−0.1155=0.8845 ⇒
σ
=Φ (0.8845)=1.2

⇒ 20−μ=1.2σ

⇒ μ+1.2σ =20......................................................................................................(i)

Page 6 of 8
 10−μ
Pr (X<10)=Φ ( )
σ
=0.0589

10−μ −1 10−μ
⇒ =Φ ( 0.0589) ⇒ = −Φ−1(1−0.0589)=−Φ−1(0.9411)=−1.56
σ σ

⇒ 10−μ= −1.56σ

⇒ μ−1.56σ =10....................................................................................................(ii)
Solving equations (i) and (ii) simultaneously, we get
μ=15 . 652 and σ=3. 623

EXERCISE
1. Let X be normal with mean = 80 and variance = 9. Find:
(a) Pr ( X >83) [0.1587] (b) Pr ( X <81) [0.6306]
(c) Pr ( X <80) [0.5] (c) Pr ( 78≤X <82 ) [0.4950]
2. Let X be normal with mean = 14 and variance = 4. Determine the constant c such
that:
(a) Pr ( X ≤c )=95 % [17.29] (b) Pr ( X ≤c )=5 % [-17.29]

(c) Pr (−c≤ X≤c )=99 % [19.152]

3. If the lifetime X of a certain kind of automobile battery is normally distributed
with a mean of 4 years and a standard deviation of 1 year, and the manufacturer
wishes to guarantee the battery for 3 years, what percentage of the batteries will
he have to replace under the guarantee? [16%]
4. A manufacturer produces airmail envelopes whose weight is normal with mean
μ=1 . 950 grams and standard deviation σ =0 . 025 grams .
The envelopes are sold in lots of 1000. How many envelopes in a lot will be
heavier than 2 grams? [About 22]
5. Specifications for a certain job call for bolts with a diameter of 0 .280±0 . 002 cm .
If the diameters of the bolts made by some manufacturer are normally
distributed with mean μ=0 . 279 and standard deviation σ =0. 001 , what
percentage of these bolts will meet specifications? [84%]
6. The lengths of bolts produced in a certain factory may be taken to be normally
distributed. The bolts are checked on two “go-no-go” gauges so that those
shorter than 2.983 in. or longer than 3.021 in. are rejected as “too-short” or “too-
long” respectively.
(i) A random sample of N = 300 bolts is checked. If they have mean length
3.007 in. and standard deviation 0.011 in., what values would you expect
for n1, the number of “too-short” bolts and n2, the number of “too-long”
bolts? [n1 = 4.4, n2 = 30.6]

Page 7 of 8
 (ii) A sample of N = 600 bolts from another factory is also checked. If for this
sample we find n1 = 20 and n2 = 15, find estimates for the mean and
standard deviation of the lengths of these bolts.[ μ = 3.001, σ = 0.01]
7. In a photographic process, the developing time of prints may be looked upon as
a random variable having the normal distribution with μ=15 . 40 seconds and
σ =0 . 48 second. Find the probabilities that the time it takes to develop one of
the prints will be:
(a) At least 16.00 seconds; [0.0668]
(b) At most 14.20 seconds; [0.0062]
(c) Anywhere from 15.00 to 15.80 seconds. [0.5934]

8. A random variable has a normal distribution with σ =10 . If the probability

that the random variable will take on a value less than 82.5 is 0.8212, what is the
probability that it will take on a value greater than 58.3? [0.9332]

9. If the probability is 0.20 that a certain bank will refuse a loan application, use the
normal approximation to determine (to three decimals) the probability that the
bank will refuse at most 40 of 225 loan applications. [0.227]
10. Show that if X is a random variable having the Poisson distribution with the
parameter λ and λ→∞ , then the moment – generating function of
X−λ
Z=
√λ
that is, that of a standardized Poisson random variable, approaches the
moment – generating function of the standard normal distribution.

11. How many workers have a salary between $ 4000 and $ 6500, if the arithmetic
mean is $ 5000, standard deviation is $ 1000 and number of workers is 15,000, if
the salary of the worker is assumed to follow a normal distribution? [11618]

12. The average daily sales of 500 branch offices was $ 150 thousand and the
standard deviation $ 15 thousand. Assuming the distribution to be normal,
indicate how many branches have sales between:
(a) $ 120 thousand and $ 145 thousand [174]
(b) $ 140 thousand and $ 165 thousand. [295]
13. As a result of tests on 20, 000 electric fans manufactured by a company, it was
found that lifetime of the fans was normally distributed with an average life of
2040 hours and standard deviation of 60 hours. On the basis of the information,
estimate the number of fans that is expected to run for:
(a) More than 2150 hours and [672]
(b) Less than 1960 hours. [1836]

Page 8 of 8