Normal Distribution

• Normal Distribution is the most useful theoretical

distribution for continuous variables.
• Many statistical data concerning business and
economic problems can be displayed in the form
of normal distribution.
• It is the most important probability model in
statistical analysis.
 Example
A company manufactures screws of 2.54 cm
diameter. There may be some variation in the
measurement. But most of the screws have
diameter very near to 2.54 cm. If we draw a
histogram of the frequency distribution of
measurements, we see that the histogram has the
maximum height at the centre and height
decreases on either side in almost same rate. So if
the class interval of the frequency distribution is
made smaller and smaller we get a curve of the
shape of a bell. It is the normal curve.
 Definition
A continuous random variable X is said to follow Normal
Distribution if its probability density function is:

Where μ is mean and σ is standard deviation. and e are

constants. The variable X varies between –ά and +ά
 Probability density function of Normal distribution

If X is a continuous random variable following normal

distribution with mean μ and standard deviation σ, then its
probability density function is given by:

Parameters of normal distribution are mean and standard

deviation. For various values of mean and standard
deviation we get different normal curves.
The equation of the normal curve gives the ordinate of the
curve corresponding to any given value of x.
Properties of Normal distribution and normal curve
1. Normal distribution is a continuous probability
distribution. The variable distributed according to
normal curve is a continuous one.
2. Normal distribution is a symmetrical distribution.
3. It is an arrangement of data set in which most values
cluster in the middle of the range and the rest taper
off symmetrically towards either extreme.
4. It is a continuous distribution of data that has the
shape of a symmetrical bell curve if it is shown
graphically. It is also known as Gaussian Distribution.
5. Probability of area above the normal curve is 0.5 and below
the normal curve is 0.5
6. Area under the normal curve will be:
Μ+/- 1σ covers 68.27% area
Μ+/- 2σ covers 95.45% area
Μ+/- 3σ covers 99.73% area
7. With a given number of σ and µ, only one normal curve
can be constructed.
8. There is only one maximum point. Hence, normal curve
is unimodel. ie. It has only one mode.
9. The curve is asymptotic to the base line. It continuous to
approach but never touches the base line.
10. The total area under the normal curve represents 100%
of individual item of the distribution.
11. The height of the normal curve is at its maximum at the
mean. Hence, the mean and mode of the normal
distribution coincide. Thus, for an normal distribution
mean, median and mode are all
equal.(Mean=median=mode)
12. The ordinates divide the area under the curve into two
equal parts. The individual values are equally distributed
on both sides of the maximum ordinate(mean)
13.Normal curve is mesokurtic. That is measure
of kurtosis is 3 (Neither platykurtic(<3) nor
leptokurtic(>3))
14.Theoretically the range of the normal curve is
between –ά and +ά. But practically the range
is μ+3σ to μ-3σ
15.Q1 and Q3 are equidistant from median
 Standard Normal Variate
If X is a random variable following normal distribution
with mean µ and standard deviation σ then variable Z=
(x-µ)/ σ is known as standard normal variate.
The Z follows normal distribution with mean 0 and SD=1.
The distribution of Z is known as standard normal
distribution.
The probability function of Z
 Understand the Standard Normal distribution and its
connection to all other Normal distributions
• A value, x, from a normal distribution specified by a
mean of μ and a standard deviation of σ can be
converted to a corresponding value, z, in a standard
normal distribution with the transformation z=(x-μ)/σ.
• In reverse, any value from a standard normal graph,
say z, can be converted to a corresponding value on a
normal distribution with a mean of μ and a standard
deviation of σ by the formula x= μ+(z*σ).
• The standard normal distribution has a mean of 0 and
a standard deviation of 1, i.e., μ =0, σ=1.
• The ability to carry out this transformation is very
important since we can do all our analysis with the
standard normal distribution and then apply the results
to every other normal distribution. For example, to draw a
normal curve with a mean of 10 and a standard deviation
of 2 (μ=10, σ=2), draw the standard normal distribution
and just re-label the axis.
• Each value along the x-axis represents that many standard
deviations from the mean. The 1 (or -1) x-value is one
standard deviation from the mean. Similarly, the 3 (or -3)
represents three standard deviations from the mean.
0
 Y
0.45

0.4

0.35

0.3

0.25

Y
0.2

0.15

0.1

0.05

0
-5 -4 -3 -2 -1 0 1 2 3 4 5
• The same concept applies to any normal
distribution, not just the standard normal
distribution. The numbers on the x-axis represent
the standard deviations from the mean. So, if the
mean is 10 and the standard deviation is 2, one
standard deviation from the mean (±1SD) yields the
values 12 and 8 (10±1*2). Similarly, ±3s yields the
values 4 and 16.
 Use of Normal Distribution
• The various tests of significance like t test, F
test, etc are based on the assumption that the
parent population from which the samples
have been drawn follows normal distribution.
• It is extremely used in large sampling theory
to find the estimates of parameters from
statistics, confidence limits, etc.
 Standard normal curve
• The equation of the normal curve gives the ordinate of the
curve corresponding to any given value of x.
• However, we are usually interested in areas under the
normal curve instead of its ordinate. The area under the
curve gives us the proportion of the cases falling between
two numbers or the probability of getting a value between
two numbers.
• The equation of normal curve depends on mean and SD and
for different values of mean and SD we will obtain different
curves. This would necessitate separate tables of normal
curve areas for each pair of values of mean and SD and
infinite number of tables required.
This problem is solved by standardizing the data
and only one table will be needed. We will be
able to determine normal curve areas
regardless of mean(μ) and SD by tabulating
only the area under the normal curve having
mean(μ) =0 and SD=1.
Such a normal curve with 0 mean and one
standard deviation is known as the standard
normal curve.
 A normal curve with mean μ and Standard Deviation
σ can be converted into a standard normal
distribution by performing the change of the scale
and origin. In the original scale (X scale) the mean
and standard deviation are μ and σ; in the new scale
(z scale) they are 0 and 1.
The formula that enables us to change x scale to z
scale and vice versa is: Z= x-µ
σ
The transformation from x to z is named as z
transformation and has the effect of reducing x to
units in terms of Standard Deviation.
Area under standard normal curve
• The standard normal curve requires the
computation of an index, which shows the
number of standard deviation that a particular
value lies to the left and right of the mean.
• The scale to which the standard deviation is
attached is known as z scale
Z= x-µ
σ
The area under this curve is equal to 1. The
curve is also called the standard probability
curve. As X increases more and more, y
becomes smaller and smaller without ever
becoming equal to zero. This means that the
curve approaches the horizontal axis but
never touches it.
Standard Normal Distribution Table
This is a table showing the probability for Z taking values
between 0 and a given value. The probability thus
obtained is the area of the standard normal curve
between the ordinate at Z=0 and the given value.
For example: mean is 1000, x is 1117 and SD is 100
Z= x-µ
σ
Z= (1117-1000)/100= 1.17
When Z is 1.17, table value is 0.3790. This table value is
the area between 0 and 1.17. This can be written as
P(0<z<1.17)=0.3790.
It may be noted that for both negative and positive values
of z, area is the same because the normal curve is
symmetrical.
i.e.; P(0<z<1.17)=P(-1.17<z<0)=0.3790
Table does not contain entries corresponding to negative
values of z. But, since the curve is symmetrical, we can
find the area between Z=0 and Z= a negative value (say
-1.17) by looking at the area corresponding to Z= a
positive value (+1.17).
 Qn
1. Find the area under the normal curve for Z=1.54
2. Also find the area under the normal curve for Z=-1.54
Answer:
1. If we look to the table, the entry corresponding to
z=1.54 is 0.4382 and this measures the area between
z=0 and z=1.54
2. Since the curve is symmetrical, we can find the area
between Z=0 and Z=-1.54 by looking the area
corresponding to Z =1.54. Hence, the area between
Z=0 and Z=-1.54 is 0.4382
Finding area under normal curve to the right of
the positive value of z
If we wish to find the area under normal curve to
the right of the positive value of z, we should
subtract the tabular value from 0.5000. The
reason is that the normal curve is symmetrical.
The area to the right of the mean is 0.5000 and
the area to the right of a positive value of z is
0.5000 minus the tabular value given for z.
 Eg:
Find the area to the right of Z=0.25
The entry in the table for Z=0.25 is
0.0987.
Subtract 0.0987 from 0.5000.
The answer will be 0.4013
Finding area under normal curve to the left of
the positive value of z
• If we wish to find out the area to the left of a
positive value, we add 0.5000 to the tabular value
given for z.
• Find the area to the left of Z=1.96
Table value for Z=1.96 is 0.4750
Area to the left of +1.96 would be
0.4750+0.5000=0.9750
 Finding the area between two values of Z, if both are
on the same side of a mean
In such case the area between them is the
difference of their tabular values.
For example, the area between z=0.60 and 1.80 is
0.2384 as given below:
Area for z value 1.80 =0.4641
Area for z value 0.60 =0.2257
the area between z=0.60 and 1.80 is 0.4641 -
0.2257 i.e. 0.2384
Finding the area between two values of Z, if
both are on the opposite side of a mean
• If the two Z’s are on the opposite side of the
mean the area between them is given by the
sum of their values.
• For example z=-0.4 and z=0.6 is
(0.1554+0.2257)= 0.3811
At times we may be given the area under the
normal curve and asked to determine the
corresponding values of Z.
For example, we may wish to find a z which is
such that the area to its right is 0.2000 (20%).
Z value corresponding to this is 0.3000 and the
closest value in the table is Z=0.84
 Ex 1
The variable X follows normal distribution with
mean 45 and SD 10. Find the probability that
i) X>60
ii) 40<X<56
iii) 45<X<60
iv) 35<X<40
 Answer
i) P(X is greater than 60)
P(X>60) : X is on the right side of the mean.
Hence, subtract the area of the Z value from 0.5000)
X=60, μ=45, σ =10
Z=(60-45)/10=1.5
P(X>60) =P(Z>1.5)
=0.5000-0.4332=0.0668
ii) P(40<X<56)
μ=45, σ =10
X lies between 40 and 56 (Both are on the opposite
side of the mean. Hence, find the sum of areas
of both the Z values)
When X= 40, Z is 40-45/10=-0.5
When X= 56, Z is 56-45/10=+1.1
P(40<X<56)= P(-0.5 <Z<1.1)
=0.1915+0.3643=0.5558
iii) P(45<X<60)
μ=45, σ =10
X lies between 45and 60. One value is
mean, hence find the area for Z value for
second value 60)
When X= 45, Z is 45-45/10=0
When X= 60, Z is 60-45/10=+1.5
P(45<X<60)=P(0<Z<1.5)
0.4332
iv) P( 35<X<40)
μ=45, σ =10
X lies between 35and 40 (i.e. Both are on the same side
of a mean. Hence, find the difference of areas of both
the Z values)
When X= 35, Z is 35-45/10=-1
When X= 40, Z is 40-45/10=-0.5
P(35<X<40)=P(-1<Z<-0.5)=
0.3413-0.1915=0.1498
 Ex 2
The scores of students in a test follow Normal distribution
with mean=80 and SD=15. A sample of 1000 students
has been drawn from the population. Find
(1) Appropriate number of students scoring between 65
and 95
(2) The probability that a randomly chosen student has
scores greater than 100. Also find Appropriate
number of students scoring greater than 100
 Answer
• μ=80, σ =15
1) Appropriate number of students scoring between 65 and 95
• P(65<X<95)
• When X=65, Z=(65-80)/15=-1
• When X=95, Z=(95-80)/15=1
• P(65<X<95)= P(-1<Z<1)
• 0.3413+0.3413=0.6826
• Appropriate number of students scoring between 65 and 95
• 0.6826*1000=683
2) The probability that a randomly chosen student has scores
greater than 100
μ=80, σ =15
P(X>100)
When X =100, Z=100-80/15 =1.33
P(X>100)= P(Z>1.33)
0.5000-0.4082=0.0918
Appropriate number of students scoring greater
than 100 is 1000*0.0918=92
 Illustration
• Assume the mean height of soldiers to be
68.22 inches with a variance of 10.8 inches.
How many soldiers of a regiment of 1000
would you expect to be over 72 inches?
 Answer
Z= x-µ
σ
µ=68.22
σ= Sqrt of variance= sqrt of 10.8=3.286
=(72-68.22)÷3.286
=1.15
=area as per table value at 1.15=0.3749
=Area where soldiers fall above 72 inches=0.5000-0.3749
=0.1251
Number of soldiers of 1000 over 72 inches= 1000*0.1251=125
 Ex 3
• The height of the school children of one
institution is normally distributed with mean
of 54 inches and SD of 12 inchers. What
percentage of students have height between
46 and 56 inches.
 Ex 4
The per acre yield of crop in a particular area is
observed to follow a normal distribution with
mean 15 Quintals and SD of 5 quintals. Find
i) The proportion of the area yielding at least 25
quintals
ii) What extent of the land under the crop can
yield between 10 and 20 quintals if the total
land under crop is 782 acres.
 Ex 5
Find the probability that the number of heads
lie in the range 185 and 220 when a fair coin is
tossed 400 times.
 Answer
• P=1/2
• n=400
• Mean = np=400*1/2=200
• SD= Sqrt of npq= Sqrt of (400*1/2*1/2)=10
• When X is 185, Z is -1.5
• When X is 220, Z is 2
• P(-1.5<Z<2)=P(-1.5<Z<0)+ P(0<Z<2)
• 0.4332+0.4772=0.9104
• Probability that the number of heads range between 185
and 200 is 0.9104
To locate values of Z when the area or probability is known

• Let ‘a’ be the unknown z value

1) If P(0<z<a)=0.4332, then how to find a?
Method: it can be seen from the table that
when table value is 0.4332, Z=1.5.
• 2) If P(z>a)=0.35, how to find a?
Method:
P(0<z<a)=0.5000-0.3500=0.1500
From the table when table value (area) is 0.1500, z
(ie. a) is 0.39
It may be noted that for finding z values the area
from 0 to a is necessary. That is P(0<z<a) is
required to find a.
 Ex 6
• In a competitive examination, 5000 students
have appeared for a paper in statistics. Their
average mark was 62 and SD was 12.
• If there are only 100 vacancies, find the
minimum marks that one should score in
order to get selected.
 Answer
Μ=62, σ =12
Total 5000 students
Vacancies 100
Percentage of students who will pass.100/5000=2%
P(z>a)= 0.02
P(0<z<a)=0.5000-0.0200=0.4800
When table value is 0.4800, a is 2.05. Therefore, z is 2.05
Using the formulae, z is (x-μ)/σ
2.05=(x-62)/12
X=(2.05*12)+62
X= 86.6 marks.
Minimum mark one should secure to get selected.
 Ex 7
The weekly wages of 1000 workmen are normally distributed
around a mean of Rs 70 and with a standard deviation of Rs 5.
Estimate the number of workers whose weekly wages will be:
i) Between Rs 70 and 72
ii) Between Rs 69 and 72
iii) More than Rs 75
iv) Less than Rs 63
Also estimate the lowest wages of the 100 highest paid workers.
 Ex 8
• Given a normal distribution with mean=40
and SD=10. Find the values of X that has
• a) 15% of the area to its left
• b) 20% of the area to its right.
 Answer
Mean=40, SD is 10
a) 15% of the area to its left
Given area is 15%
Let x=a
Then P(X<a)=15%=0.15 OR P(z<a)
Then P(mean>X>a)=0.5000-0.1500=0.3500 P(0>z>a)
When x=a, Z=(x-μ)/σ
From the table, z value corresponding to this area 0.35 is 1.04
Take negative value of 1.04 i.e. Z=-1.04 as it is left of the mean.
-1.04=(X-40)/10
-10.4+40=X
X=29.6
 Answer..continued..
Mean=40, SD is 10
b) 20% of the area to its right
Given area is 20%
Let x=b
Then P(X>b)=20%=0.20 OR P(z>b)
Then P(mean<X<b)=0.5000-0.2000=0.3000 P(0<z<b)
When x=b, Z=(x-μ)/σ
From the table, z value corresponding to this area 0.30 is 0.85
Take positive value of 0.85 i.e. Z=-0.85 as it is right of the mean.
0.85=(X-40)/10
8.5+40=X
X=48.5
 Ex 9
• In a normal distribution 17% of the items are
below 30 and 17% of the area above 60. Find
the mean and SD
 10
The following table gives frequencies of occurrence of
variate X between certain limits.
Variate(x) frequency(f)
Less than 40 30
40 or more but less than 50 33
50 and more 37
100
The distribution is exactly normal. Find the average and SD.
 Ex 11
• In a certain examination, the percentage of passes and
distinctions were 46 and 9 respectively. Estimate the
average marks obtained by the candidates, the
minimum pass and distinction marks being 40 and 75
respectively. Assume the distribution of marks to be
normal.
• Also determine what would have been the minimum
qualifying marks for admission to a re-examination of
the failed candidates, had it been desired that the 25%
of them should be given another opportunity of being
examined.
 Answer
Let ‘x’ denoted the marks obtained by the students.
i) Pass indicates X>=40
P(X>=40)=46%=0.46
P(0<z<a)=0.50-0.46=0.04
Z value for 0.04 is 0.1
But, z=(x-μ)/σ
0.1=(40- μ)/σ
0.10σ+μ=40....(1)
 Answer...contd..
ii) Distinction indicates X>=75
P(X>=75)=9%=0.09
P(0<z<a)=0.50-0.09=0.41
Z value for 0.41 is 1.34
But, z=(x-μ)/σ
1.34=(75- μ)/σ
1.34σ+μ=75....(2)
• 0.10σ+μ=40....(1)
• 1.34σ+μ=75....(2)
1.24 σ=35
σ=28.22
μ=37.18
b) Since the percentage of failed students is 54%
(100-46=54).
25% of failed students should be re-examined, ie, 25% of
54=13.5%
Let z value be c when x is x3
P(c<z<a)=0.135
P(c<z<0)=0.135-0.04=0.095
Z value corresponding to the area 0.095 is -0.24 (take
–value since the area is left of the mean.
z=(x- μ)/σ
-0.24=(x- μ)/σ
-0.24=(x- 37.2)/28.22
X=30.43
 Ex 12
• A student secures 72 marks in an examination in
sociology for which his class average is 54 and the
SD is 20. He secures 76 marks in Statistics for
which his class average is 52 and the SD is 12.
• What can you say about the performance of this
student with reference to these two
examinations. (Convert to Z scores for both
examination and make your comment).
 Ex 13
• The income distribution of workers in certain
factory was found to be normal with mean
Rs 1000 and SD Rs 100. There were 180 persons
getting above Rs 1200. How many persons were
there in all.
 Fitting Normal curve.
• Main problem in fitting the normal distribution is
estimating the parameters of the distribution.
Parameters of the distribution are μ and σ. So
find the mean and SD of the given frequency
distribution. Then we get the probability density
function. By substituting value of μ and σ in the
equation, we get the equation of the best fitting
normal distribution.
 Steps to get theoretical frequencies
1) Calculate mean and SD of the given distribution
2) Calculate Z values for all class limits, using the formula, Z=(x-
μ)/σ
3) Find the area of each Z value from the Standard Normal Table
4) Find the area of each class by subtracting or adding, as the
case may be, with the areas obtained for z values
5) Multiply the area of each class by the total frequency to get
the class frequency
The new frequency distribution with theoretical frequencies, will
be a normal approximation to the given distribution.
 Ex 14
Fit a normal distribution to the following data:
Marks: No of students
10-20 4
20-30 22
30-40 48
40-50 66
50-60 40
60-70 16
70-80 4
Marks: No of Mid (X) d’ d’2 fd’ fd’2
(Class) students
(f)
10-20 4 15 -30 900 -120 3600
20-30 22 25 -20 400 -440 8800
30-40 48 35 -10 100 -480 4800
40-50 66 45 0 0 0 0
50-60 40 55 10 100 400 4000
60-70 16 65 20 400 320 6400
70-80 4 75 30 900 120 3600
200 -200 31200
=45+(-200/200)
= 45-1=44

=SD= 12.5
The density function is:
 (1) (2) (3) (4) (5)= (4)*200
Class limit Z=(x-44)/12.5 Areas from table Area for interval Theoretical
of X frequency

10 -ά 0.5000
20 -1.92 0.4726 0.0274 5
0.1040 21
30 -1.12 0.3686 0.2431 49
40 -0.32 0.1255
50 +0.48 0.1844 0.3099 62
0.2153 43
60 +1.28 0.3997 0.0815 16
70 +2.08 0.4812
80 +ά 0.5000 0.0188 4
1.
Various Steps
First find mean and SD of the given distribution i.e.μ and σ
2. Then for each X value (class limit), get z value using the
formula, Z=(x- μ)/ σ =(x- 44)/ 12.5. These Z values form column 2
and area for these Z values form column 3
3. Colum 4 is the difference between two adjacent values of column
3 except two values where z changes from negative to positive.
4. Column 5 is obtained by multiplying the values in column 4 with
total frequency 200.
Note: first and last z values are taken as – ά and + ά.
In column 4, forth value (0.3099) is obtained by adding two
values of 3 column as the values changes from negative to
positive.
 Ex 15
• Fit a normal curve to the following data.
Marks: No of students
5-9 10
10-14 22
15-19 40
20-24 21
25-29 7
 Answer
2 2
Marks: No of Mid value d’ d’ fd’ fd’
students

4.5-9.5 10 7 -10 100 -100 1000

9.5-14.5 22 12 -5 25 -110 550
14.5-19.5 40 17 0 0 0 0

19.5-24.5 21 22 5 25 105 525

24.5-29.5 7 27 10 100 70 700

100 -35 2775

=17+ (-35/100)
= 17-0.35=16.65

=SD= 5.26
The density function is:
 (1) (2) (3) (4) (5)= (4)*100
Class Z=(x-16.65)/5.26 Areas Area for interval Theoretical
limit of X from table frequency

4.5 -ά 0.5 … ….
9.5 -1.36 0.4131 0.087 9 9
0.254 2525
14.5 -0.41 0.1591
19.5 0.54 0.2054 0.365 3636

24.5 1.49 0.4319 0.227 2323

0.068 7 7
29.5 +ά 0.5 … ….
 Ex 15
• Fit a normal curve to the following data.
Diameter Frequency
7 1
8 6
9 7
10 11
11 20
12 10
13 6
14 5
15 1
Diameter Frequency
6.5-7.5 1
7.5 -8.5 6
8.5-9.5 7
9.5-10.5 11
10.5-11.5 20
11.5-12.5 10
12.5-13.5 6
13.5-14.5 5
14.5-15 1
 Ex 16
Marks obtained by a number of students are
assumed to be normally distributed with mean
50 and SD 5. If 6 students are taken at
random, what is the probability that exactly two
of them will have marks over 60.
 56
In an examination, it is laid down that a student passes if
he secures 30% or more marks. He is placed in the
first, second or third division as he secures between
60% and 80% marks, between 45% and 60% marks and
marks between 30% and 45% respectively. He gets
distinction in case he secures 80% or more marks. It is
noticed from the results that 10% of students failed in
the examination whereas 5% of them obtained
distinction. Calculate the % of students placed in the
second division (Assume normal distribution). Also
Calculate the % of students placed in the first division
Answer
1) 10% of students failed in the examination
Pass marks : 30
P(X<30)=10%=0.10
P(30<x<Mean)=0.50-0.10=0.40
Z value for the area 0.40= -1.28 (Take negative value
since the value falls on the left side of the mean)
Z=(x-μ)/σ
-1.28= (30- μ)/σ
-1.28 σ+ μ=30.................(1)
2) Student gets distinction in case he secures 80% or
more marks
• 5% of them obtained distinction.
• P(X>80)=0.05
• P(Mean<X<80=0.50-0.05=0.45
• Z value for the area 0.45 is 1.65 (Take + value as it falls
on the right side.
Z=(x-μ)/σ
1.65= (80- μ)/σ
1.65 σ+ μ=80.................(2)
• -1.28 σ+ μ=30.................(1)
• 1.65 σ+ μ=80.................(2)
Solving the above two equations ((2)-(1)) we get:
2.93σ=50
σ=17.06
μ=51.85
% of students placed in the second division
(scored marks between 45% and 60% marks)

P(45<X<51.85) plus P(51.85<X<60)

1) P(45<X<51.85)
• Z=(x-μ)/σ
• Z=(45-51.85)/17.06
• -0.4015
• Area for the Z value 0.4015=0.1591
2) P(51.85<X<60)
Z=(60-51.85)/17.06
=0.477
Area for the Z value 0.477=0.1808
Probability that student scored marks between 45% and
60% marks=0.1591+0.1808=0.3399=34%
2) % of students placed in the first division
(scored marks between 60% and 80% marks)
• P(mean<X<60)=18.08%
• P(X>80)=5%
• P(60<X<80)=50-(18.08+5)=26.92%