Normal Distribution

Normal Distribution: A random variable X whose probability density function is given by

1 𝑥−𝜇 2
1 − ( )
𝜑(𝑥) = 𝑒 2 𝜎 , −∞ < 𝑥 < ∞ (1)
𝜎 √2𝜋

is called a normal variate. We also say that X is normally distributed with parameters 𝜇 and 𝜎 2 , and is
denoted as 𝑿~𝑵(𝝁, 𝝈𝟐 ). The continuous distribution given by eq. (1) is called the normal distribution.

Remark 1. The cumulative distribution of 𝑋~𝑁(𝜇, 𝜎 2 ) is denoted as ∅(𝑥). Thus

𝑥
∅(𝑥) = 𝑃[𝑋 ≤ 𝑥] = ∫−∞ 𝜑(𝑢)𝑑𝑢.

∞
Remark 2. It can be verified that ∫−∞ 𝜑(𝑥)𝑑𝑥 = 1.

1 𝑥−𝜇 2
1 − ( )
Remark 3. The curve 𝑦 = 𝜎 2𝜋 𝑒 2 𝜎 is called the normal curve which possesses the following
√
properties:

Fig. 1.1
i. The curve is bell shaped and symmetrical about the line 𝑥 = 𝜇.
1
ii. The maximum ordinate is 𝜎 , which occur at 𝑥 = 𝜇.
√2𝜋

iii. Mean, median and mode of the normal distribution coincide i.e. at 𝑥 = 𝜇.

(𝑋−𝜇)
Standard Normal Variate: If 𝑋~𝑁(𝜇, 𝜎 2 ), then 𝑍 = 𝜎
is called the standard normal variate with

𝐸[𝑍] = 0 𝑎𝑛𝑑 𝑣𝑎𝑟[𝑍] = 1, and is written as 𝑍~𝑁(0,1).

1
1 − 𝑧2
The p.d.f. of standard normal variate Z is given by 𝜑(𝑧) = 𝑒 2 , −∞ <𝑧<∞
√2𝜋

𝑧
and the corresponding distribution function is given by ∅(𝑧) = 𝑃[𝑍 ≤ 𝑧] = ∫−∞ 𝜑(𝑧)𝑑𝑧.

∞ 1 ∞ −1𝑧 2
Remark: Since ∫−∞ 𝜑(𝑧)𝑑𝑧 = 1, therefore, ∫ 𝑒 2 𝑑𝑧 = 1.
√2𝜋 −∞

1 2
∞
Hence, ∫−∞ 𝑒 −2𝑧 𝑑𝑧 = √2𝜋 .
 1 2 2
Theorem: If 𝑋~𝑁(𝜇, 𝜎 2 ), then 𝐸[𝑋] = 𝜇, 𝑣𝑎𝑟[𝑋] = 𝜎 2 𝑎𝑛𝑑 𝑀𝑋 (𝑡) = 𝑒 𝜇𝑡+2𝜎 𝑡
.

∞
Proof: The m.g.f. of X is given by 𝑀𝑋 (𝑡) = 𝐸[𝑒 𝑡𝑋 ] = ∫−∞ 𝑒 𝑡𝑋 . 𝜑(𝑥)𝑑𝑥

2
1 ∞ 𝑡𝑋 −1(𝑥−𝜇) 1 ∞ 𝑡{𝜇+𝜎𝑧} −1𝑧 2 𝑥−𝜇
𝑀𝑋 (𝑡) = 𝜎 √2𝜋
∫−∞
𝑒 .𝑒 2 𝜎 𝑑𝑥 = ∫−∞
𝑒 . 𝑒 2 𝑑𝑧, where 𝑧 = ( )
√2𝜋 𝜎

∞ ∞
𝑒 𝑡𝜇 1 2 𝑒 𝑡𝜇 1 2 −𝜎2 𝑡 2 }
𝑀𝑋 (𝑡) = ∫ . 𝑒 −2(𝑧 −2𝑡𝜎𝑧) 𝑑𝑧 = ∫ . 𝑒 −2{(𝑧−𝜎𝑡) 𝑑𝑧
√2𝜋 √2𝜋
−∞ −∞

1
𝑡𝜇+ 𝜎2 𝑡2 1
𝑒 2 ∞ − (𝑧−𝜎𝑡)2
𝑀𝑋 (𝑡) = ∫−∞. 𝑒 2 𝑑𝑧. (Put 𝑢 = 𝑧 − 𝜎𝑡)
√2𝜋

1
𝑡𝜇+ 𝜎2 𝑡2 1 2 1 2 2 1 2 2
𝑒 2 ∞ − 𝑢 𝑡𝜇+ 𝜎 𝑡
𝑀𝑋 (𝑡) = ∫−∞. 𝑒 2 𝑑𝑢 = 𝑒 2 × 1 = 𝑒 𝑡𝜇+2𝜎 𝑡
. (1)
√2𝜋

1 2 2
Hence, 𝑀𝑋 (𝑡) = 𝑒 𝑡𝜇+2𝜎 𝑡
is the required m.g.f. of the normal distribution.

Differentiating (1) with respect to 𝑡, we get

1 2 2
𝑀𝑋′ (𝑡) = 𝑒 𝑡𝜇+2𝜎 𝑡
(𝜇 + 𝑡𝜎 2 ) ⟹ 𝐸[𝑋] = 𝑀𝑋′ (𝑡)|𝑡=0 = 𝜇 (2)

Differentiating (2) with respect to 𝑡, we get

1 2 2 1 2 2
𝑀𝑋′′ (𝑡) = 𝑒 𝑡𝜇+2𝜎 𝑡
(𝜇 + 𝑡𝜎 2 )2 + 𝑒 𝑡𝜇+2𝜎 𝑡
. 𝜎2

𝐸[𝑋 2 ] = 𝑀𝑋′′ (𝑡)|𝑡=0 = 𝜇2 + 𝜎 2 . (3)

Hence, 𝑣𝑎𝑟[𝑋] = 𝐸[𝑋 2 ] − (𝐸[𝑋])2 = 𝜇2 + 𝜎 2 − 𝜇2 = 𝜎 2 .

𝑏−𝜇 𝑎−𝜇
Theorem: If 𝑋~𝑁(𝜇, 𝜎 2 ), then 𝑃[𝑎 < 𝑋 < 𝑏] = ∅ ( 𝜎
)− ∅(𝜎
). (Try yourself)

𝑏
Theorem: If 𝑋~𝑁(0,1), then 𝑃[𝑎 < 𝑋 < 𝑏] = ∅(𝑏) − ∅(𝑎) = ∫𝑎 ∅(𝑧)𝑑𝑧. (Try yourself)

𝑧
Area under a normal curve: The definite integral 𝛹(𝑧) = ∫0 𝜑(𝑧)𝑑𝑧 is called the normal probability
integral, which gives the area under the standard normal curve between the ordinates 𝑍 = 0 𝑡𝑜 𝑍 = 𝑧. Also
due to symmetry, it gives the same area under the ordinates 𝑍 = −𝑧 𝑡𝑜 𝑍 = 0.

𝑧 1 𝑧 −1𝑧 2
The value of ∫0 𝜑(𝑧)𝑑𝑧 = ∫0
𝑒 2 𝑑𝑧 for different values of 𝑧 at the interval of 0.01 are given in the
√2𝜋

following table:
Important Results:

1. 𝑃[−𝑧1 ≤ 𝑍 ≤ 𝑧1 ] = 2𝑃[0 ≤ 𝑍 ≤ 𝑧1 ] = 2𝑃[−𝑧1 ≤ 𝑍 ≤ 0].

2. 𝑃[−𝑧1 ≤ 𝑍 ≤ 0] = 𝑃[0 ≤ 𝑍 ≤ 𝑧1 ].
3. 𝑃[𝑍 ≤ −𝑧1 ] = 𝑃[𝑧1 ≤ 𝑍].
4. 𝑃[𝑍 ≤ 𝑧1 ] = 0.5 + 𝑃[0 ≤ 𝑍 ≤ 𝑧1 ].
5. 𝑃[−𝑧1 ≤ 𝑍] = 0.5 + 𝑃[−𝑧1 ≤ 𝑍 ≤ 0].
6. 𝑃[0 ≤ 𝑍 ≤ 𝑧1 ] = 0.5 𝑎𝑛𝑑 𝑃[−𝑧1 ≤ 𝑍 ≤ 0] = 0.5 𝑤ℎ𝑒𝑛𝑒𝑣𝑒𝑟 𝑧1 ≥ 2.5 𝑎𝑝𝑝𝑟𝑜𝑥.

###: While solving problems, we are generally given that X is a normal variate with mean
𝜇 and standard deviation 𝜎. We should convert the variate X into the standard normal
𝑋−𝜇
variate Z by means of the transformation 𝑍 = and further the given probability under
𝜎

the normal curve should be converted to the form 𝑃[0 ≤ 𝑍 ≤ 𝑧] in order to make use of
the normal table.
Example 1. If X is a normally distributed with mean 2 and variance 1, find 𝑃[|𝑋 − 2| < 1].
𝑋−𝜇
Solution: We have 𝜇 = 2 𝑎𝑛𝑑 𝜎 = 1. Let 𝑍 = = 𝑋 − 2.
𝜎

Now, 𝑃[|𝑋 − 2| < 1] = 𝑃[2 − 1 < 𝑋 < 2 + 1] = 𝑃[1 < 𝑋 < 3] = 𝑃[1 − 2 < 𝑋 − 2 < 3 − 2]

⟹ 𝑃[|𝑋 − 2| < 1] = 𝑃[1 − 2 < 𝑋 − 2 < 3 − 2] = 𝑃[−1 < 𝑍 < 1]

⟹ 𝑃[|𝑋 − 2| < 1] = 2𝑃[0 < 𝑍 < 1]

⟹ 𝑃[|𝑋 − 2| < 1] = 2 × 0.3413 = 0.6826 from table.

Example 2. If X is a normal variate with mean 50 and variance 100, find 𝑃[𝑌 ≤ 3137],

𝑤ℎ𝑒𝑟𝑒 𝑌 = 𝑋 2 + 1 𝑎𝑛𝑑 𝛹(0.6) = 0.2258 .

Solution: Given that 𝜇 = 50 𝑎𝑛𝑑 𝜎 = 10.

Therefore, 𝑃[𝑌 ≤ 3137] = 𝑃[𝑋 2 + 1 ≤ 3137] = 𝑃[𝑋 2 + 1 ≤ 3136 + 1]

𝑃[𝑌 ≤ 3137] = 𝑃[𝑋 2 ≤ 3136] = 𝑃[𝑋 2 ≤ 562 ] = 𝑃[−56 ≤ 𝑋 ≤ 56]

−56 − 50 𝑋 − 50 56 − 50
𝑃[𝑌 ≤ 3137] = 𝑃 [ ≤ ≤ ] = 𝑃[−10.6 ≤ 𝑍 ≤ 0.6]
10 10 10
𝑋−𝜇
𝑃[𝑌 ≤ 3137] = 𝑃[−10.6 ≤ 𝑍 ≤ 0] + 𝑃[0 ≤ 𝑍 ≤ 0.6], by putting 𝑍 = 𝜎

𝑃[𝑌 ≤ 3137] = 0.5 + 𝛹(0.6) = 0.5 + 0.2258

𝑃[𝑌 ≤ 3137] = 0.7258.

1
Example 3. If 𝑙𝑜𝑔𝑒 𝑥 is normally distributed with mean as unity and variance 4, find 𝑃 [ < 𝑥 < 2]. Given
2

that 𝑙𝑜𝑔𝑒 2 = 0.693.

1 1
Solution: 𝑃 [2 < 𝑥 < 2] = 𝑃 [𝑙𝑜𝑔𝑒 2 < 𝑙𝑜𝑔𝑒 𝑥 < 𝑙𝑜𝑔𝑒 2] = 𝑃[−0.693 < 𝑙𝑜𝑔𝑒 𝑥 < 0.693]

1 1
𝑃 [ < 𝑥 < 2] = 𝑃 [𝑙𝑜𝑔𝑒 < 𝑙𝑜𝑔𝑒 𝑥 < 𝑙𝑜𝑔𝑒 2] = 𝑃[−0.693 < 𝑙𝑜𝑔𝑒 𝑥 < 0.693]
2 2
1
𝑃 [2 < 𝑥 < 2] = 𝑃[−0.693 < 𝑦 < 0.693], where 𝑦 = 𝑙𝑜𝑔𝑒 𝑥 ~ 𝑁(1,4)

1 −0.693 − 1 𝑦 − 1 0.693 − 1 𝑦−𝜇 𝑦−1

𝑃 [ < 𝑥 < 2] = 𝑃 [ < < ] , 𝑡𝑎𝑘𝑒 𝑍 = = ~ 𝑁(0, 1)
2 2 2 2 𝜎 2
1
𝑃 [2 < 𝑥 < 2] = 𝑃[−0.85 < 𝑍 < −0.15] = 𝑃[0.15 < 𝑍 < 0.85], by symmetry

1
𝑃 [2 < 𝑥 < 2] = 𝛹(0.85) − 𝛹(0.15) = 0.3023 − 0.0596 = 0.2427.
Example 4. The local authorities of Chapra installed 2000 electric lamps in the street of city. If the lamps
have an average life of 1000 burning hours with a S.D. of 200 hours. Assumes that the lives of the lamps
are normally distributed. (i) What number of the lamps might be expected to fail in the first 700 burning
hours, (ii) after what period of burning hours would we expect that (a) 10% of the lamps would have failed
(b) 10% of the lamps would be still burning?

Solution: (i) Let X denote the life of lamps, then 𝑿~𝑵(𝟏𝟎𝟎𝟎, 𝟐𝟎𝟎𝟐 ).

𝑋−1000 700−1000 𝑋−𝜇 𝑋−1000

𝑃[𝑋 < 700] = 𝑃 [ < ] = 𝑃[𝑍 < −1.5], where 𝑍 = =
200 200 𝜎 200

𝑃[𝑋 < 700] = 𝑃[𝑍 > 1.5] = 0.5 − 𝑃[0 ≤ 𝑍 ≤ 1.5] = 0.5 − 0.433, from table

𝑃[𝑋 < 700] = 0.067.

Thus the number of lamps expected in first 700 hours of burning = 2000 × 0.067 = 134.

(ii) (a) Let 𝑡 denote the period in hours when 10% of the lamps would still be burning. Then

𝑃[𝑋 < 𝑡] = 0.1 ⟹ 𝑃[𝑋 ≤ 𝑡] = 1.0 − 0.1 = 0.9

𝑋−1000 𝑡−1000 𝑡−1000

⟹ 𝑃[ 200
≤ 200
] = 0.9 ⟹ 𝑃 [𝑍 ≤ 200
] = 0.9

𝑡 − 1000
⟹ 𝑃 [0 ≤ 𝑍 ≤ ] = 0.9 − 0.5 = 0.4 = 𝑃[0 ≤ 𝑍 ≤ 1.28]
200
𝑡−1000
⟹ 200
= 1.28 ⟹ 𝑡 = 1256. (From table)

Hence 10% of the lamps would still be burning

after 1256 burning hours.

(b) If we draw a figure, it follows from the

figure that 10% of the lamps would have failed
after 1000-256 = 744 burning hours

Example 5. Of a large group of men, 5% are under 60 inches in height and 40% are between 60 and 65
inches. Assuming a normal distribution, find the mean height and standard deviation.

Solution: If 𝑿~𝑵(𝝁, 𝝈𝟐 ), then we are given 𝑃[𝑋 < 60] = 0.05 𝑎𝑛𝑑 𝑃[60 < 𝑋 < 65] = 0.40.

⟹ 𝑃[𝑋 < 60] = 0.05 𝑎𝑛𝑑 𝑃[𝑋 < 65] = 0.05 + 0.40 = 0.45.

𝑋−𝜇 60−𝜇 𝑋−𝜇 65−𝜇 𝑋−𝜇

⟹ 𝑃[ 𝜎
< 𝜎
] = 0.05 𝑎𝑛𝑑 𝑃 [ 𝜎
< 𝜎
] = 0.45. (Put 𝑍 = 𝜎
)
 60−𝜇 65−𝜇
⟹ 𝑃 [𝑍 < ] = 0.05 𝑎𝑛𝑑 𝑃 [𝑍 < ] = 0.45.
𝜎 𝜎

60−𝜇 65−𝜇
⟹ 𝑃[ < 𝑍 < 0] = 0.5 − 0.05 𝑎𝑛𝑑 𝑃 [ < 𝑍 < 0] = 0.5 − 0.45.
𝜎 𝜎

60−𝜇 65−𝜇
⟹ 𝑃[ 𝜎
< 𝑍 < 0] = 0.45 𝑎𝑛𝑑 𝑃 [ 𝜎
< 𝑍 < 0] = 0.05. (due to symmetry)

𝜇−60 𝜇−65
⟹ 𝑃 [0 < 𝑍 < 𝜎
] = 0.45 𝑎𝑛𝑑 𝑃 [0 < 𝑍 < 𝜎
] = 0.05. (By table)

𝜇−60 𝜇−65
⟹ 𝑃 [0 < 𝑍 < 𝜎
] = 𝑃[0 < 𝑍 < 1.645] 𝑎𝑛𝑑 𝑃 [0 < 𝑍 < 𝜎
] = 𝑃[0 < 𝑍 < 0.13].

𝜇−60 𝜇−65
⟹ 𝑃 [0 < 𝑍 < 𝜎
] = 𝑃[0 < 𝑍 < 1.645] 𝑎𝑛𝑑 𝑃 [0 < 𝑍 < 𝜎
] = 𝑃[0 < 𝑍 < 0.13].

𝜇−60 𝜇−65
⟹ 𝜎
= 1.645 𝑎𝑛𝑑 𝜎
= 0.13.

⟹ 𝜇 = 65.42 𝑎𝑛𝑑 𝜎 = 3.29.

Exercises:

1. X is a normal variate with mean 30 and S.D. 5. Find the probability that:
(i) 26 ≤ 𝑋 ≤ 40, (ii) 𝑋 ≥ 15 and (iii) |𝑋 − 30| > 5. Ans. 0.7653, 0.00135 and 0.3174.
2. If 𝑙𝑜𝑔10 𝑋 ~N(4, 22) and 𝑙𝑜𝑔10 1202 = 3.08, 𝑙𝑜𝑔10 8318 = 3.92,
then find 𝑃[1.202 < 𝑋 < 83180000]. Ans. 0.95
3. Let X be the life in hour of a radio tube. Assume that X is normally distributed with mean 20 and
variance 𝜎 2 . If a purchaser of such radio tubes requires that at least 90% of the tubes have lives
exceeding 150 hours. What is the largest value of 𝜎 can be and shall have the purchaser satisfied?
Ans. 132 approx. Hint: (P[X>-150]≥0.90)
4. Suppose that the diameter of the shafts manufactured by a certain machine are normal random variable
with mean 10 cms and S.D. 0.1 cm. If for a given application the shaft must meet the requirement that
its diameter fall between 9.9 and 10.2 cms, what proportion of the shafts made by this machine will
meet the requirement? Ans. 81.85%
5. If the skulls are classified as 𝐴, 𝐵 𝑎𝑛𝑑 𝐶 according as the length-breadth index is under 75, between 75
and 80, and over 80. Find approximately the mean and S. D. of a series in which A are 58%, B are 38%
and C are 4%. (Assume that the distribution is normal and use table). Ans. 74.4 and 3.2
6. Show that for a normal distribution 𝛽1 = 0 𝑎𝑛𝑑 𝛽2 = 3.
7. Fit the normal curve to the data (means find 𝜇, 𝜎 𝑎𝑛𝑑 𝑛𝑜𝑟𝑚𝑎𝑙 𝑝. 𝑑. 𝑓.):
x (mid-point) 2 4 6 8 10
f (frequency) 1 4 6 4 1
Ans. 𝑀𝑒𝑎𝑛 = 6 𝑎𝑛𝑑 𝑆. 𝐷. = 2.