NORMAL DISTRIBUTION

Introduction
❖ A continuous r.v X that can assume any value within a specified interval of values.
❖ The r.v X is said to be normally distributed with mean µ and standard deviation  if its
probability distribution is given by:

1 (𝑥−𝜇)2
1 − ( )
𝑓 (𝑥 ) = 𝑒 2 𝜎 , for -∞ <𝑥< ∞
𝜎 √2𝜋

❖ If X is a continuous r.v, then it can be written as

𝑋 ~ 𝑁(𝜇, 𝜎 2 )
Properties of Normal Distribution
❖ The curve is bell-shaped and has a single peak.
❖ The distribution is symmetrical about 𝜇
❖ Total area under the curve is equal to 1. The mean is at the middle and divides the area into
halves (± 0.5)
❖ The curve never touches the x axis.

Shapes of The Normal Distribution

𝑁(𝜇, 𝜎 2 )

𝜇
Standard Normal Distribution
❖ The probability for normal distribution for a particular interval [a,b] is given by:
𝑏

𝑃(𝑎 < 𝑋 < 𝑏) = ∫ 𝑓(𝑥)𝑑(𝑥)

𝑎

❖ Since it is hard to integrate, STANDARD NORMAL TABLE is used.

❖ The standard normal distribution has all the properties of a normal distribution with µ=0 and
𝜎2 = 1
𝑍 ~ 𝑁(0,1)
❖ To convert other normal distributions to standard normal distribution, then this formula is
used:
𝑋−𝜇
𝑍=
𝜎
❖ Normal distribution tables : the probabilities are in the form P(Z > k)
Example 1
If Z ~ N (0,1), find from the standard normal table:

a) P(Z > 1.54) = 0.0618

b) P(Z < 0.65) = 1 – P(Z > 0.65)

= 1 – 0.2578

= 0.7422

c) P(Z > -0.95) = P(Z < 0.95)

= 1 – P(Z > 0.95)

= 1 – 0.1711

= 0.8289

d) P(Z < -0.72) = P(Z > 0.72) = 0.2358

e) P(1.22 < Z < 2.40) = P(Z > 1.22) – P(Z > 2.40)
= 0.1112 – 0.0820
= 0.103

f) P(-0.75 < Z < 1.29) = P(Z > -0.75) – P(Z > 1.29)
= P(Z < 0.75) – P(Z > 1.29)
= [1 – P(Z >0.75)] – P(Z > 1.29)
= [1 – 0.2266] – 0.0985
= 0.6749

g) P(0 < Z < 0.4) = P(Z > 0) – P(Z > 0.4)

= 0.5 – 0.3446
= 0.1554

h) P(-0.6 < Z < 0) = P(Z > -0.6) – P(Z > 0)

= P(Z < 0.6) – P(Z > 0)
= [1 – P(Z >0.6)] – P(Z > 0)
= [1 – 0.2743] – 0.5
= 0.2257

Example 2
If Z ~ N (0,1), find from the standard normal table:

a) P(Z > 1.36) e) P(Z > -1.27)

b) P(Z ≤ 1.36) f) P(-1.99 < Z < -0.99)
c) P(0.34 ≤ Z ≤ 1.41) g) P(-1 < Z < 0.57)
 d) P(Z ≤ -1.36) h) P(2.13 < Z < -1.37)

Note:

1) P(Z < k) = < 0.5 → k = -ve value

2) P(Z < k) = > 0.5 → k = +ve value

3) P(Z > k) = > 0.5 → k = -ve value

4) P(Z > k) = < 0.5 → k = +ve value

Example 3
If Z ~ N (0,1), find the value of k if:

a) P(Z < k) = 0.28 d) P(Z > k) = 0.1562

k = -0.58 k = 1.01

b) P(Z < k) = 0.975  k − 15.3 

e) P Z   = 0.28
 0.05 
1- P(Z > k) = 0.975

P(Z > k) = 0.025 k − 15.3

= -0.58
0.05
k = 1.96
k – 15.3 = -0.029

k = 15.271
c) P(Z > k) = 0.9821

1- P(Z < k) = 0.9821

P(Z < k) = 0.0179

k = -2.1

Example 4
If Z ~ N (0,1), find the value of k if:

a) P(Z > k) = 0.15

b) P(Z < k) = 0.08
c) P(Z > k) = 0.78
d) P(Z < k) = 0.90
  k − 9.88 
e) P Z   = 0.20
 0.12 

Example 5
If X ~ N (51,22), find:

60.5 − 51
a) P(X ≥ 60.5) = P( Z  ) = P( Z  2.03) = 0.0212
22

47.6 − 51
b) P(X < 47.6) = P( Z  ) = P( Z  −0.72) = P( Z  0.72) = 0.2358
22

44 − 51 54 − 51
c) P(44 < X ≤ 54) = P( Z ) = P(−1.49  Z  0.64)
22 22
= P( Z  −1.49) − P( Z  0.64)
= [1 − P( Z  −1.49)] − P( Z  0.64)
= [1 − P( Z  1.49)] − P( Z  0.64)
= [1 − 0.0681] − 0.2611
= 0.6708

Example 6
If Y ~ N (299,24), find from the standard normal table:

a) P(Y > 307)

b) P(Y < 312)
c) P(290 < Y < 299)

Probabilities [Problems Involving the Normal Distribution]

Example 7
A machine produces components with a mean diameter 15.35 and a standard deviation of 0.05.
The diameters are assumed to be normally distributed.
a) Find the probability that a randomly chosen component has a diameter:
i) less than 15.29 mm
ii) between 15.35 and 15.43 mm
iii) more than 15.37 mm
b) If components with diameters outside the range between 15.28 and 15.40 mm are rejected,
what percentage of these components will be rejected?
c) If 28% of these components have diameters less than a value k, find k.

Solution:

X ~ N (15.35, 0.0025)

 x −  15.29 − 15.35 
a) i) P(X < 15.29) = P   = P(Z  −1.2)
  0.05 
= P(Z > 1.2)
= 0.1151
  15.35 − 15.35 x −  15.43 − 15.35 
ii) P(15.35 < X < 15.43) = P   
 0.05  0.05 
= P(0 < Z < 1.6)
= P(Z > 0) – P(Z > 1.6)
= 0.0452 – 0.5
= 0.4452

 x −  15.37 − 15.35 
iii) P(X > 15.37) = P   = P(Z > 0.4) = 0.3446
  0.05 

b) P(component will be rejected)= P( X  15.28)  ( X  15.40)

 x −  15.28 − 15.35   x −  15.40 − 15.35 
= P   + P  
  0.05    0.05 
= P(Z < -1.4) + P(Z >1)
= P(Z > 1.4) + P(Z > 1)
= 0.0808 + 0.1587
= 0.2395

c) P(X < k) = 0.28

 x −  k − 15.35 
P   = 0.28
  0.05 
 k − 15.35 
P Z   = 0.28
 0.05 
k − 15.35
= 0.28
0.05
k = 15.36 mm

Example 8
The average amount of water from a dispenser machine is 7 gm per glass with standard deviation
of 0.5 gm. A glass would be overfilled if it has more than 7 gm of water. It is known that the amount
of water is normally distributed.
i. What is the probability that a glass has more than 7.8 gm of water?
ii. What is the probability that a glass has 6.7 gm to 7.3 gm of water?
iii. Find the probability that a glass of water will be overfilled?
iv. If there are 100 glasses, determine the number of glasses that would be overfilled.

Example 9
The amount of fruit juice in a box is normally distributed with a mean 1010 ml and standard
deviation 8 ml.
i. Find the probability that a certain box contains more than 1004 ml fruit juice.
ii. 8% of the boxes were rejected since they contained too much juice. Find the maximum
amount of juice in a box so that it is not rejected.
iii. The label at the boxes indicates that each box contains 1 liter of fruit juice. If there are 2000
boxes, find the number of boxes which have juice less than the amount stated on the label.