Module ll

Statistical Techniques-I

CHAPTER Statistical Techniques

INTRODUCTION
Statistical Techniques. The methods based on statistics are known as statistical technigues

5.2 STATISTICS
Statistics is a branch of science dealing with the collection of data, organising, summarising.
presenting and analysing data and drawing valid conclusions and thereafter making reasonabie
decisions on the basis of such analysis.

53 FREQUENCY DISTRIBUTION
or
Frequency distribution is the arranged data, summarised by distributing it into classes
categories with their frequencies.
Wages of 100workers
Wages in 0-10 10 - 20 20-30 30-40 40-50
Numbers of workers 12 23 35 20 10

54 GRAPHICAL REPRESENTATION
It is often useful to represent frequency distribution by means of a diagram. The diferent
types of diagrams are
1. Histogram 2. Frequency polygon
3. Frequency curve 4. Cumulative frequency curve or Ogive
5. Bar chart 6. Circles or Pie diagrams.
a
1. Histogram consists of set of rectangles having their heights proportional to the
class-frequencies, for equal class-intervals. For unequal class-interval, the areas of rectangles ar
proportional to the frequencies.

36 36

workers 30 workers
30
24 24
18 18
of
No. 12
of
No. 12 --
6 6

X 0 10 20 30 40 50
0 10 20 30 40 50
Histograrm 2024/1O/17 22:46 Frequency polygon

144
 Statistical Techníques 145

2.Frequency Polygon is a line graph of class-frequcncy plotted against class-mark. It can be

btaiped by connecting mid-points on the tops of the rectangles in the histogram.
3. Cumulative Frequency curve or the Ogive. If the various points are plotted according
as x co-ordinate and the cumulative frequency as y c0-ordinate and
the upper limit of the class
hoce noints are joined by a frec hand smooth curve, thc curve obtaincd is known as cumulative
frequency curve or the Ogive.

ERMEASURES OF CENTRAL TENDENCY OR AVERAGES

An average is a value which is representative of a set of data. Average value may also be
termed as measure of central tendency. Therc are fivc types of common measures of central tendency.
(i) Arithmetic average or nean
(ii)Median (iii) Mode
(n) Geometric Mean (v) Harmonic Mean

5.6 ARITHMETIC MEAN

IfX,, K,,Ny ...,are n numbers, then their arithmetic mean (A.M.) is defined by.
n

..(1)
If the number x, occurs ff times, x, occurs f, times and so on, then

AM= .2)
5+t....*t.#,
(a) Direct method, uses above two formulae.
Example 1 Find the mean of 20, 22, 25, 28, 30.

Solution. A.M. =
20+ 22+25+28+30 125
=25 Ans.
5 5
Example 2. Find the mean of the following
Number 8 10 15 20
Frequency 5 8 8 4
Solution. Efr = (8 x 5)+ (10 x 8) + (15 x 8) + (20 x 4)= 40 + 80 + 120 + 80 = 320
Ef=5 + 8+ 8 + 4 25
=

320 =
A.M. = 12.8 sLante a Ans.
25
(b) Short cut method
Let a be the assumed mean, d the deviation of the variate x from a. Then
Sfu-a) = A.M.– =A M -4 as

A.M. = a+

xAmple 3,
Find the arithmetic mean of the following distribution:
Class 0- 10 10- 20 20- 30 30- 40 40- 50
Frequency 7 8 20 10 51:
 - IV
Introduction to Engineering Mathematics
E146
Solution. Let assumed mean (a) = 25.
Class Mid-value x Frequency f x- 25 = d fd
5 7 -20 l40
0- 10
s 10 - 20 -80
25 20
20- 30
- 40 10 +10 +100
30 35
+20
40 - 50
5 +100
45
Ef= 50 Zfd
Total =0
A.M. = a+ = 25 + -20 -= 24.6 Ans,
50

(c) Step deviation method

Let a be the assumed mean, i the width of the class interval and

X-a
Sfd,
d= A M.=a+
i

Example4. Find the arithmetic mean of thedata given in example 3 by step deviation metho
Solution. Let a = 25

a
Class Mid-value x frequencyf d
=

fd.
0- 10 7 -2 -14 9t.
10 - 20 15 8 -1
20-30 25 20
30 - 40 35 10 + | + 10
40 -50 45 +2 + 100

Total
2f= 50
2fd = -2

A.M. = a+ ()=25+ -x10= 24.6

50
5.7 MEDIAN
Median is defined as measure of
the the central item when they are ascending o
descending order of magnitude. arranged in
When the total number of the items is odd
and equal to sav n then the value of (77
item gives the median.
When the
2
total number of the items even, say and
so
is n, then there are two middle items,
the mean of the values of nth and
items is the median.
Example 5. Find the median of
.
6,
8,9, 10, 11, 12, 13.
Solution. Total number of items =7

The middle item = = 4th

2024/10/17 22:46
 iM Statistícal Techniques 147

Median =Value of the 4th item = 10 Ans.

N
For grouped data, Median l+| 2

where / is the lower limit of the median class, the frequency of the class, i is the width of the
fis
class-interval, C is the cumulative frequcncy of the class preceding the median-class and N is total
frequency of the data.
Example 6
Find the value of Median from the following data:
No. of days for which absent (less than) 5 10 15 20 25 3) 35 40 45
No. of students 29| 224 465| 582 634 644 650 | 653 655
Solution. The given cumulative frequency distribution will first be converted into ordinary
frequency as under

Class- Interval Cumulative frequency Ordinary frequency

0-5 29 29 = 29

5- 10 224 224 - 29 = 195

10 - 15 465 465
–
224 =241
15 -20 582 582 - 465 = 117
20 - 25 634 634 - 582 = 52
25-30 644 644 - 634 = 10
30 - 35 650 650 - 644 = 6
653 - 650 = 3
35 – 40 653
40 -45 655 655 - 653 = 2
655
Median=size of
2
Or 327.5th item
327.5th item lies in 10-15 which is the median class.

Nc
M= 1+2

where I stands for lower limit of median class,

N stands for the total frequency,
C stands for the cumulative frequency of the class preceding the median class,
istands for class interval
J Stands for frequency for the median class.
655 - 224
2
Median= 10+ 241

= 10+ 103.5 ×5
10 + 2.15 = 12.15 Ans.
241
2024/1O/17 22:46
148 o Introduction to Engineering Mathematics -IV

5.8 MODE
variable which
occurs most frequently.
Mode
is definedto bethc sizc of'the
Dxample 7.
Find the mode of the following items :
2, 6, 0, 5, 6, 0. or more than 5.
times, hence the
0, 1, 6, 7, 2, 3. 7. 6. 6. occurs 5

As 6 occurs 5 timcs and no other

item
Solution, An,
mode is 6.

For grouped data, Mode =l+|

2f- L-h, is the
nodal class, fis the frequency of the modal class, i
where 7 is the lowcr limit ofthe is the frequency after the modl
of the class,f, is the
frequency before the modal class and
Emperical formula -Median]
Mean - Mode =3 [Mean
8. Find the mode from the following
data:
Example
12- 18 18- 24 24- 30 30-36 36-42
Age 0-6 6-12
25 35 18 12 6
Frequency 6

Solution.
o Age Frequency Cumulative frequency

0-6 6 6
6-12 11 17
12 - 18 25 -f1 42
|18–24 35 =f 77
24-30 18
=f 95
30-36 12
107
36- 42 6
113
Mode =
1+-
(3f-f-5)
= 18 + 35- 25
70- 25-18)|x6
= 18+ 60
c=18+2.22 =
5.9 GEOMETRIC
MEAN 27 20.22 Ans

Ilfx, X,, Xy
s,x, be n values of variates x,
then the
geometric mean
Example 9.
Find the geometric
Solution. GM.= mean
(4x8x16)' =8. of4, 8, 16.

5.10 HARMONIC MEAN Ans

Harmonic mean
their reciprocals. of a series
Thus if H of values
be the harmonicis defined as
mean the reciprocal mean l
ther of the arithmetic
 Statistical Techniques 149

+....t
Example 10. Calculate the harmonic mean of4, 8, 16.

Solution. 7
H 48
48
H=: =6.853 Ans.
7

5.11 PARTITION VALUES

These are the values which divide the series into a number
of equal parts.
are denoted by 2,
quartile
, and Q. The first quartile ,
The three points which divide the series into four equal parts are called

, Quartiles. They
exceeds 25% of the observations, the second
coincides with the median and the third quartile O, exceeds 75% of the observations.
The nine points which divide the series into ten equal parts are called Deciles and are denoted
by D,, D,, ., D:
The ninety nine points which divide the series into hundred equal parts are called Percentiles.
They are denoted by P, Pz .,Pg- The method of computation of partition values are the
same as those of median in case of both discrete and continuous distributions.
aieil
Formulae for Quartiles, Deciles and Percentiles:
-c
. =l+ i. r=1, 2, 3

D, =l+| 10 i. r=1, 2,... 9

xr-c
100
P=l+| |, r=1, 2, 3,...99

5.12 MEASUREs OF DISPERSION

Averages discussed earlier fail to reveal the Full details of the distribution. Two or Three
distributions may have the same average but still they may differ from each other in many ways.
In such cases, further statistical analysis of the data is necessary so that these differences between
Various series can be studied and accounted for. Such analysis will make our results more accurate
and we shall be more confident of our conclusions.

Suppose, there are three series of nine items each as follows:

As
Series Series B
VSeries Css
50 48 5

50 50 15

50 46 20

2024/10/1722-46
 Introduction to Engineering Mathematics -V
T50
25
50 49
35
50 47
80
50 52
85
50 53
90
51
50 95
50 54
450
Total 450 450
50
Mean 50 50
of
identical. The items are not
all the items is
In the first series, the
mean is
50 and the value However, in the second case,
representative of this distribution.
at
all scattered, and the mean is the series have different values.
But the items are
not
mean is 50 yet all the items of the is 54 in
though the series is 46 and the
maximum therange.
as the minimum value of the
&
between
very much scattered the diference
mean a good representative of the series because vabs
also, the mean is 50 and the mesn
is
In this case also, very significant. In the third series
Imean and other items is not values are very widely scattered. Though the
are also different, but here the fomation
of different items yet the series differ widely fom each other in their
is the same in all the three
series,
in this group and
to
satisfactorily represent the individual items
Obviously, the average does not among the items in the
completely, further analysis is essential. The scatter
know about the series a small range, while in the third
case the values
case it varies within
first case nil, in the second It is evident from the above. that
a

span and they are widely scattered.

range between a very big on the
extent the scatter around average should also be made to throw more light
study of the of
to this scatter is dispersion.
composition of a series. The name given
Definition
Some important definitions.of dispersion are given below:
or variation of the variable about a
(i) "Dispersion or spread is the degree of the scatter
central value." Brooks and Dick
(ii) "Dispersion is the measure of the variations of the items.'" -A. L. Bowley
(ii) "The degree to which numerical data tend to spread about an average value is called
the variation or dispersion of the data." Spiegel

(i) "Measures of variability are usually used to indicate how tightly bunched the samp
values are around the mean." Dyckman and Thomas
From the above definitions, it is clear that in a general sense the term dispersion refers to
ul
varíability in the size of items. If the variation is considerable
substantial, dispersion is said to be
and if the variation is very little, dispersion
is insignificant.
Different Measures of Dispersion
Measures of Dispersion [or Variability]

Absolute
measures
Relative
Range Inter-quartile
Quartile
measuresO
range Average Standard
deviation deviation deviation Lorenz
[Mean Curve
deviation]
Coefficient Coefficient
Coefficient Coefficient
of range of mean variation
of quartile of
2024/10O 17 2 deviation deviation
 Statistical Techniques
*
5.13 RANGE
Range is the simplest measure of dispersion. It is the difference between two extreme
observations of the distribution. :
312fi:1Ets*
Range =L-S
here L is the largest value and S is the smallest value in a series.
With all its limitations, Range is commonly used in
Quality Control, Weather forecasting,
ariation in money, Sales, Share Values, Gold prices etc.

5.14 QUARTILE DEVIATION

Inter Quartile Range

Q.D. = ,-2
where , 2 2
% are the first and third quartiles of the distribution. It is an absolute measure of
and
dispersion. Ifit is divided by the average value of the two quartiles, a relative measure
of dispersion
is obtained.

Coefficient of Q.D.

5.15 AVERAGE DEVIATION OR MEAN DEVIATION

It the mean
is of the absolute values of the deviations of a given set of numbers from their
arithmetic mean.

lf zy X ,
be a set of numbers with frequencies f,
the arithmetic mean of the numbers x, X, ...., x, then
f, ...
f, respectively. Let I be

Mean deviation =

Example t1. Find the mean deviation of the following frequency distribution.

Class 0-6 6-12 12-18 18- 24 24-30

Frequency 10 12 9 5

Solution. Let a = 15
Class Mid- valuex Frequencyf| d
=afd
0-6 3 8 -12 -96 11 88
6-12 9 10 60 5 50
12-18 15 12 0 1
12
18-24 21 9 +6 54 i3363
24-30 27 + 12 60 13 E65
Total Lfd =-42 278
Ef= 44 f -14|=
42 =14 (nearly)
Mean ()=a+: =15-
44

/|*-_278
Average deviation = 44
=6.3 Ans.
 Introduction to Engineering
Mathematics- IV
1523
STANDARD DEVIATION
5.16 as the squarc root of the mean of the
square of the deviation
S Standard deviation is defined
from the arithmetic mean.

=
ru-)²
S.D. G=

deviation g² is called variance.

Note. 1. The square of the standard by u

is called the second moment about the mean and is denoted

2. o'

CALCULATING STANDARD DEVIATION

SHORTEST METHOD FOR
5.17
We know that o'-fa-f-s-a-z-a
-fd--a Where I-a d

N
=
1
-2(7- a) Efa +
(F- a³ (:: y=N]
N N
1 1

I=a+ I-a=
N N

S.D, O=
N

Note. Coefficient of variation =-*100

Example 12, Calculate the mean and standard deviation for the following data:
Size of item 7 9 10 12
Frequency 3 13 8 5 4
Solution. Assumed mean (a)=9

dx- a
fd.
3 -3 -9 27
7 6 •-2 -12 24
8 9 -1 -9 9
9 13
10 +1 8
2024/10/17 22:47
 Statistical Techniques A53
11 5 +2 10 |20
12 4 +3 12 36
Sf- 48 fA = 124

Mean = a+
=9+0=9istl al)

S.D.= |124
V
=1.6 Ans.
48
Example 13 From the following frequency distribution, compute the standard
:
deviation of 100 students
Mass in kg bt 60 - 62 63 - 65 66- 68 69 -71 72 - 74
Number of students 18 42 27
Solution. Assumed mean (a) = 67
Mass in kg Number of students
f d= x-67 f.d|
60-62 5 61 -6 -30 180
63 - 65 18 64 -3 -54 162
66 –68 42 67 0
69 -71 27 70 3 81 243
72-74 73 6 48 288
fa = 873
f- 100 fd= 45

|873
S.D.= V100

= V8.73-0.2025=V8.5275 =
2.9202
Ans.
Example 14, Compute the standard deviation for the following frequency distribution:
Class interval 0-4 4- 8 8- 12 12 - 16
Frequency 4 2
Solution. Assumed mean
Class interval d=x--6 fd
0–4 4 2 -4 - 16 64
4-8 6 0 o
8- 12
- 16
10 32i
12 1
14 + 8 64
f= 15 fd=0 Sfa = l60

=
160
S.D. -0 =3.266 Ans,

2024/10/17 22:47
 Introduction to Engineering Mathematics - IV
154
5.18 MOMENTS
mathematics, a moment is a
Moment is a very brief period time, whereas in of specific
a
quantitative measure of the shape of function. measure a
a set of statistical
parameters to distribution.
In statistics, moments are said to be
we apply the formulae Ld
Calculating mcan deviation and standard deviation n
and
On,

deviation from the actual

mean and n
is the total number
where d is the

moment about the mean Similatly

of items. are also called First Moment and Second

momnent are calculated as and Ifthe deviation

third moment, fourth moment and h n
a
are not taken from the actual arithmetic mean but from (assumed mean),the irst moment, second
a ara
moment, third moment, fourth moment and moment about any number n

E(r-a) E(-a)'E(r-a)' E(r-a) E(x- a)'

and

5.19 MOMENTS ABOUT MEAN

Let be the arithmetic mean, then

L=Sa-), r= 0,1, 2, 3, 4...... where, N=>f

i=l
i=l

If r= 0,
H(;-=1
N =1

If r= 1,
Nj=l S;-)=0
Ifr=2
N i=1 lu, = variance]
If r = 3,

Ifr=4,
N
i=l
520 MOMENTS ABOUT ANY NUMBER
(RAW MOMENTS)

K,--a)',
N i=l
r= 0,1,2, .. =
where N
If r=0, -a
Kof(a, =1

If r= 1, n

2d/no/7 22:48
 od3ApniStatistical Techniques

If r= 2, N
1

If r=3,
i=1

If r=4, N

S21 MOMENTS ABOUT THE ORIGIN

v,
Nl
=1
Ifr= 0,
=fx
Vo

i=1
n

Ifr= 1, v
N i=l
1

If r= 2,

If r=3,
i=1
1 n
If r= 4, V47
i=l

5.22 RELATION BETWEEN L, AND u,

We have,

N i=l
- a) - (F-a)]
-SN i=l

On expanding by Binomíal theorem,

+"C,u-.. +(-1y
H,
="C
2024/10722,43
156Introduction to Engineering Mathematics - IV

Putting r=2,3, 4 we get

Hy
=',- 2(1",+ (u', ()=-
Hs
=-
=y-3u',
3(',) ', (u';)'- (
,
+ 3

+2 ()

Thus, we have thc following relations:

H
='4- 4u',u' +6 -3
on the R.H.S. is zero.
Note. 1. The sum of the coefficients of the various terms
2. The dimension of each term on R.H.S. the same as that of terms on the L.HS
on L.H.S.
3. The number of terms on R.H.S. = order of the moment
Conversely
1 1

+ "C,
= EiG-y +"GEi,-E-) Ef6-)-a
.C%-Z)F- a)+SE-ay]
-i-)+GEf-Io- a)

N N

2
If r=1 = X -a
If r= 2,
=H, +0 +'?
If r= 3,
L4= 0

If r=4,

=a

5.23 RELATION BETWEEN v,

AND u,

,2., 3..
i1224/i,
 ensi gN StatisticalTechniques

=( -a+ a)'
On expanding by binomial theorem, we have

- a)' +
=I(,
N i=|
G (- ay-'ut ...+a]
=,+"C,ua t.... +a
*, we get
a = v,
On putting =H,+'C, H
t "C,
Hxt... x
..(1)
On taking r= 1, 2, 3, 4 in (1), we get
=
H,
=0, 4, 1]

= y
tCH7+'Czu7= +

= + +
l4 4u,F+6u,F

V3-+3u,+ V4
t4u,F+ 6u, +

E
Example 15. Find the first four moments for the following individual series:
3
12 20
Solution. Here we have, x=9
Sl. No.
(x-)' (*-F)
1
-8 64 - 512 4096
2 3 -6 36 -216 1296
3
0 0 0
4 12 3
27 81
5
20 121 1331 14641
n=5 £x= 45 Z(x-)=0Z(x-'=230 (- T)

=630E(*-F)= 20114

Ex 45
X= -=9
E(x-X)-o
Z(r-)' 230
-= 46
n 5
630
=126
5
E(r-) 20114
= 4022.8
Ans.
2824/10/A78
EM58 Introduction to Engineering Mathematics -IV
FExamplc 16 Calculate the variance and third central moment from the following datg
4 5 6. 7
2 3
9 26 59 72 52 29 ns7
(U.P III Semester Dec. 2005)

Solution.
,-4 f-4)(K,-4)f-4)9
0 -4 -4 6 64
243
81
1
-3 -27
2 26 -2 - 52 104 - 208
3 59 -1 - 59 59 59

4 72 0
5 52 1 52 52 52

6 29 2 58 116 232
7 7 3 21 63 189
1
4 4 16 64
| Ef= 256 Ef(, -4) =-7 Ef (, -4= 507
f (x,–4) =-37
– 4)
2f (x -7
256

Ef (; –4)' 507
256
-
2(% 4)' -37
256
2
507 -7
256
=1.98047 -0.00075 = 1.97972
256)
Hy +
-3u,. 2(4

-37 3
3

256
0.14453 +0.16246 – 0,00004
= 0,01789
Ans.
Example 17
Calculate u, Hz Hy Hafor
the following frequency
distribution:
| Marks 0-10 10-20 20-30 30-40 50-60
No. of students 40-50
6
15 11 7

2024/1072248
 yd Statistical Techniques 159
Solution.
No. Mid
of
students value
Marks f(*-T)4

0-10 5 5 -30 -30 900 --27000 810000

10-20 15 90 - 20 - 120 2400 -48000 960000
20-30 10 25 250 - 10 - 100 1000 - 10000 100000
30-40 15 35 525
40-50 11 45 495 110 1100 11000 11000
30-60 7 55 385 20 140 2800 56000 112000

|Ef- 50
2fr= 1750 |2f(*-)Efa-}| f(r-? zf(u-)|
=0 = 8200 =- 18000 =3100000
fr 1750
=35
Ef 50

2f(x-) 0
= -=0
Ef 50

Ef(r-z' 8200
-= 164
f 50

-18000
=-360
2f 50

Ef(r-)* 3100000
-=62000 Ans.
Ef 50
Example 18. The first three moments of a distribution, about the value 2'of the variable
are 1, 16
and– 40. Show that the nean is 3, variance is 15 and u, =- 86.
Solution. We have

a=2, =1, ,= 16 and u, =– 40

We know that
X=u+a=1+ 2
=3
Variance -4,--=16- (I = 15

-3u, u+ 2=-40 - 3(16)(1) + 2(1)³ =-40

–
Hy
48 + 2--86,.
35'are - 1.8,
SxAnple 19, The first four moments of'a distribution,
about the value 240,
-1020 and 144000, Cind
the values of l y 4
Solution. We have, a =
35,
=-1.8, H,=240, =- 1020, ,= 144000

Hy-H= = 236.76
240 --1.8)2

2024,/10/17 22:4.8
160
i* introduction to Engineering Mathematics - IV
M, -
3u, u + 2u - 1020 - 3(240)(– 1.8) + 2(- 1.8)- 2642r

= 144000 - 4(- 1020)(- 1.8) + 6(240)(- 1.8) - 3(- 1.8y

= 141290.11. Ans.
Example 20. The first three moments ofa distribution about lhe value 2 of the variable
1, I6 and- 40. Find the values of the fst three moments about the origin.
Solution. We have,
a=2, u= #= 16, n=-40 1,

We have u,=u-u= 16 -l 15 - =

4, =-3u, +
2u=-40 3(16)(1) + 2(1)³
=-40 -48 + 2=- 86

We have V, = X =a+u =2 +1=3

V, = y + = 15 + (3) = 24

=
V, M, +
3u, T+ =-86 + 3(15)X3) + (3)³ =76. Ans.
Example 21 The first four moments ofa distribution about 2 are 1, 2, 5, 2.5 and 1 6. Calculate
the first four momnents about the mean and about origin.
Solution. We have,
a=2, u=1, H=2.5, =5.5, H= 16.
Moments about mean

Hy=-(H)= 2.5 -(1 = 1.5

H
=-3u, + - 3(2.5)(1) +
2(H)= 5.5 2(1³ = 0
H=-4u, +6uH-3u= 16 -4(5.5)(1) + 6(2.5X1² – 3(1) = 6.
Moments about origin
Let v, be the rth moment
about origin. We have the
following formulae:
V= 7-a+u V,=T=2+1=3
V,=1.5 +
(3)² = 10.5
V,=0+3(1.5)(3) + (3)³ =
40.5
V4=6+4(0)(3) + 6(1.5X3)² +
(3)= 168.

2024/10,/7 22:48
 GhStatistical Techniques * 161

EXERCISE 5.1
Calculate first four moments about the mean, for the following individual series:
5 5 5 5 5 5
, =
Ans. u,
=
0

.Eind the first four moments about the mean of the following series:
10
3
,
1

Ans. = 0, , =
12, u, =- 12, H,=208.8
Find the first four moments for the following frequency distribution:
2 3 4 5 6 8
1
1
3 4 4 3 2
4, u, H,=37.6
= =
Ans. u, =0, H, 0,
A Find the first four moments about the mean for the following frequency distribution:

Mark 0-10 10-20 20-30 30-40 40-50

No. of students 5 10 40 20 25
=- 300,
, Ans. u, =0, 125, u, u, =37625
=
H,

5. Calculate ,, Has Ha for the series

:
4, 7, 10, 13, 16, 19, 22.
= 0, L, =2268
Ans. u, =0, H, = 36,
6. If the first four moments of a distribution about the value
5 are equal to – 4, 22, 117 and 560. -
Determine the corresponding moments:
(i) about the mean, and (ii) about zero Ans. (i) 0, 6, 83, 992 (ii) 7, 102, 1361
mean. Also, compute the first four moments
7. Compute first four moments of the data 3, 5,7,9 about the
about the point 4. Ans. u, =0, H, 5, H, =0,
, = 9,
=
41 ,=
',= 38, ',= 177
P=2,
5.24 GROUPING ERROR AND ITS SHEPPARD'S CORRECTIONS
(FOR MOMENTS)
symmetrical and the number of class intervals is greater than th
If the distribution is not 20
error.
Of the range, then the computation of moments will have an error known as grouping

=
H, (corrected)
12

H, (corrected) = -, 2
+: 1_
240
no correction.
Where h is the width of the class-interval while u, and 4, require
Inese formulae are known as Sheppard's corrections.

P25 MOMENT GENERATING FUNCTION (M.G.F.)

x = a is defined as the expected value
monent generating function of the variate x about
e

of er- a) and is denoted

by M, ().
M, () = Ep, e-a)

2024/10/17 22:49
162 Introduction to Engineering Mathematics - IV

En1*,-a)+
Ep, (x - ay +.
=
E, +
tp,(x, - a) + Ep, (x, -a) +...
2 r!
- a)
=
1+1(
-
+(a,
a) -a)' +...+ L, t..
2t...+ it... [2p,=|=H
wherc u, is the nmoment of order r about a.

Hencc. H= coeficient of
or
p= di
Jt=)

Again M,() = p, )
=
ep, e

= e
M, ()
Thus the moment generating function about the point a=e moment generating function
about the origin.

Example22, Find the corrected values ofthe followingmoments using Sheppard's corrections.
H; =210, =
460, H,= 96700, h = 12
Solution. We have,
= =
H, 210, H
460, H,= 96700, h= 12

H, (corrected) = , 12
=210
(12)²
12
= 210–12 =
198
H, (corrected) = 460

H,(corrected) '= 7
240
= 96700- (12)2 7
-(210) + (12)
240
= 96700 - 15120 + 604.8 = 80975.2
Curve of
data is flat topped or
platykuritc.
Example 23. Calculate Ans.
the first Four moments of the
correction. following data. Also Sheppard's
make
Values
10-20 20-30 30-40 40-50
Frequency 50-60
20 60-70 70-80
69 108
78
222

2024/10/7 22:49
 a
Statistical Techniques *

Solution.
Calculation of Moments

x-45 2
S3
Values
d=r-45 10

15 - 30 -3 3 9 27 81
10-20

20 25 - 20 -2 40
- J60 320
20-30

69 35 - ]0 -1 69 69 -69 69
3040
108 45 0
40-50

78 55 10 1
78 78 78 78
50-60

22 65 20 2 44 88 176 352
60-70

70-80 2 75 30 3 6 18 54 162

(x-45 2

N=300 10 -45 10 10
= 16 = 52 = 1062
-342

Now X-45 16
10 = 0.53
-2 10
-10=|
300

342
-10 = (10) =114
300

K--45 10
-10 = 52
300
(10) =173.33

1
x-45) 4 -10 =| 1062 =35400.
10 300
Moments about mean

H =0 (always)
Hz*-= 114- (0.53)² = 113.72
K,+ - 3(114)(0.53) + 2(0.53³
HyHj-3u, 2u= 173.33 =-7.63

=35400 – 4(173.33)(0.53) + 6(114)(0.53)? - 3(0.53) =35224,44. snoi

Sheppards Correction, =
h 10
H
(Corrècted) = , =0
(10)²
H (Corrected) = y - 12 =113.72- 12 -=105.39
Example 243
Findthe moment generating function ofBinomial distribution.
Solution. Here, we
have
f(o) ="C,p o4/10/172249
 IV
Introduction to Englneering Mathematics-

Moment generating function about the origin

M

() = E e, "C, p"
q*

+...[By Binomial Theroem)

q +
"C, qpe "C,q2 (pe'y

= lg +pe|"
M,(O) +pey'=-(ng +
pe'y.(pe')
dt dt
J=0
n (q +p-p= np

-n(g + pe y= (pe )),-0

dt
J=0
= [n -1)(g
(n
+
pey-2 (pet) +n (g +pe')y- (pe)I,.
= [n(n -1) (q +p)" -p + n (g +p".pl
n (7 - 1)p' + np Lg
+p=1]
np [(n- 1)p+ 1]=np[np + (1-p)] = np [np + q]

, =

=
n'p + npq
=V, = np

-=V, -vý (n´p“ t npq) -(np)

=
H=

Standard deviation=

Mean =
Example 25. Find the moment generating
function of the discrete Poisson distribution given by
em*
x!
Hence, find the first and
second moments about mean
Solution. Here, we have and also the value of variance.
e m

=
f()
x!
Moment generating
function about the origin

Mo ()= e"m
x!
e'y
=e
x!
(me')?
in =e"1+
me' + (me' y
2! 3!
.eme' =e(e-i)

2024/1017 2249
 i Statistical Techniques 165

=v,=
Jf=) J)

me' =D,m= m 2

J=0
d
', =V,= =
d
J=0
me
-|eme -) (me'y' +eme-D,
Jt=0

-e(l-D m'+eml-I) .m=m+m

4, = Vy-=V,-v =
(m' +
m)-m² =m
Hence, Imean =1
m
Variance= Ans.
Example 26 Find the moment genierating function of the random variable whose moments
are
1)! 2r
,=(r+
Solution. M,() =
E(e*) = (r+1)(21)
r=0

=
=0 r! r=0 r! r=0
+ ... =
=1+ 2.2t +3. (2r)? (1- 2/)2 Ans.

526 MOMENT GENERATING FUNCTION OFA CONTINUOUS VARIATE

The moment generating function of the continuous probability distribution about x = a is
given by

M, () = e- f() dr
DZ7 PROPERTY OF MOMENT GENERATING FUNCTION
Ihe moment generating function of the sum of two independent chance variables is the product
of
their
respective moment generating functions.
Oymbolically, M,.,(0) = and y are independent random variables.
M() x M,() provided thatx
roo1. Let x and y be two independent random variables so that x +y is also a random variable.
The m.g.f. of the sum x
+y w.r.t. origin is
M

Since x
) =EP, {eet) = p, {e,ey =p, (e") .Ep, (ey
andy are independent variables and so are e
and e',
Hence,
M,„()= M, (). M, (0 Proved.

2024/10/17 22:49
168 *
Introduction to Engineering Mathematics
- IV
of the exponential
distribution
function
Example 27 Find the moment generating
1
(U.P. II Semester, Dec. 2005)
f(x)=÷ee, 0<rSoo, c>0
C

deviation.
Hence, find its mean and standard
origin is
The moment generating function about the
Solution.
dx
M, =
() e.f()

=
C

1 1

=
1
C
t
C

-=(l-cr)
1- ct
=1 ctt c²+ c+ [Binomial Theorem)
+
M, ()
d +e
SA Moment about origin = -[1+ ct + c't +...0
t=0 dt
=
[c+ 2c t+ 3c Pt ...],-0=c
==c

J=0

-[1+ ct + c' ++.

d -[c
+2ct+
dt 3¢'t+......-0s1l30
- (2c' +6c't +...Je0

Standard deviation =
Ans
 .

Statistical Techniques 167

amale 28. Obtain the momentgenerating function of the random variable x having
probability distribution

for 0<x<I
s()={2- for Isx<2
0 elsewhere
Also determine u, H, and u, (G.B.T:U., III Semester Dec, 2012)

Solution. M)= Je
f() dr

=
[.ar +
(2-1) e dr t+J 0.e" dk +
2
2e2 e2r
e-2e'+1
2
21 4/2 &,3 164
...
2 6 24

41 2,2 2
-$+2+ +
3
-

3 12
+....+1

= 4 1
3
coeficient of
1
3

4 1
8-1 7
P= coefficient of 2 3 6 6 6

H*-(u=-(I²=!
6 6
Hence, u' =
1,
#=, H2
Ans.
Example 29, Find the moment generating
given by fnction of the continuous normal distribution
1

f()= o 2
262 ,-oo <.X<o

Solution. Here, we
have (*)= o e 26
2

MOment generating function about the origin

M. () = e
26 dx
o 2 ...(1)
Putting =z so that dx = o dz
in (1), we get

2024/10/722:49
 Mathematics –IV
* Introduction to Engineering
tz dz
e 2 (odz) =
e.2
M, ()=ez+#)
1

dz = dz
e e
/2

1
dz =e dzV
e 2
N2

Ans.

assuming only non-negative values has a POm

Example 30. The random variable X
distribution is given by
probability distribution if its probability
;x>0, a>0, B>1|
f (x)={rB
0, elsewhere

Find the noment generating function of Ganma probability distribution.

(x) dx
Solution. M,()= ef
TB Jo
--e(a-Ddx.

r Jo
. ye dy | where y =x (a.- ) so that dy=(a-) d:

(a-P
1

||<a.
rß

Xe p(X = x), ifx is discrete

Note: M,(t) =x= 0'

if x is continuous

EXERCISE 5.2
Find the moment generating
function of the following
functions:
1. f()= mem x,
m>0
Ans,
2. f) =
e*(1+e,- o<x<o
Ans. B
(1 -t, 1 + ),
1-t>0
3. Ifx is a discrete random Tt COsec Tt,
t<1
variable with probability
Find its moment generating function f(x) = x=1, 2....... constant)
function. (k=
Ans.
e'
2024/10 17 22050
 Statistical Techniques
o
169
Find the moment generating function for the distribution where.

at x=l
3

f(r)=1 Ans. 2
at x= 2 7et
otherwisc

5. Find the moment generating function of the random variable x having the probability density function:

for -1<Ki<2
f(r)=k Ans.
kt
0. otherwise 3
for t= 0

6 A random variable x has density function given by

x20
[0 x<0

Obtain the moment generating function. 2

Ans. if t<2
2-1
7. Find the moment generating function, if it exists, given the probability
distribution frequency :

xe
X>0
otherwise Ans.
l-t
8. Find the moment generating function for the given distribution

f(x)
for a<x<b
=b-a Ans. t(b-a)
otherwise
9. A random variable x has the probability distribution function

for x=1, 2,3........

f(x) =22 Ans.
0, otherwise 2-e
Obtain the moment generating function.

5,28 SYMMETRY
Adistribution is said to be symmetrical when its mean,
median and mode are identical. i.e.
Mean = Median =
Mode.
other a
distributed words, distribution is said to be symmetric when the frequencies are symmetrically
about
have the same the mean (or when the values of the variables are equi-distant from the mean
frequency). and
Consider
the following frequency distribution:
X 10 20 30 40 50 60 70
2 6 10 14 10 6 2
Mean n1Mode
Median
2024/10/17 22.50
 Introduction to Engineering Mathematics-V

×10) + (40x14) +(50 x10)+ (60x6) + (70x2)

(10x2) +(20 x6)+(30
Mean = T= 2+6+10-+14+10+6+2
2000 =
40
504 20 and 60 are cquidistant
fromthe mean,
In this distribution,
we obscrve thát the values viz.

40 with the same frcquency (. a perfectly bell-shaped

plotted on a graph will give curve,
A
symmetrical distribution when
which is known as normal curvc.

SKEWNESS
5.29
Skewness denotes the opposite of
symmetry. It is lack of symmetry,
(U.P. III Semester, 2006)
Skew symmetrical Distribution
to be skew symmetrical distribution. In skeu
A distribùtion which is not symmetrical said
is
are not of equal length. One tail will be lon
symmetrical distribution the left tail and the right tail
than the other.
(a) Negatively skewed distribution.
In negatively skewed distribution the longer Median

Left tail is
Mode.
Mean

left tail is longer than the right tail.

(b) Positively skewed distribution.
In positively skew distribution the
right tail of the curve will be
longer than the left.
tail Median
Right
Left tail longer
Mode
is
In skew distribution mean, median and
mode are not equal.

5.30 DEFINITION OF SKEWNESS

A distribution is said to beskewed' when the mean and median fall at different points in the
distribution and the balance (or centre of gravity) is shifted to one
side or the other to left or right -
5.31 TEST OF SKEWNESS
1 There is no skewness in the distribution
if AM = Mode = Median
2, There is no skewness in the distribution
if,
Thirdquartile - Median = Median -
First quartile.
3. There is no skewness if aya
The sum of the frequencies which are
less than = Mode
Sum of the frequencies wnl
ti:a 4. There is no skewness are greater than Mode
if quartiles are equidistant
5. The distribution from the median.
is negatively skewed
6. The curve is not if A,.M. is less than Mode.
symmetrical if AM Median
Mode.
5.32 USES OF SKEWNESS
1. It indicates whether the distribution
is normal or not.
2. It tells about the concentration
abou: irean.O/17 22S0
 aila vStatistical Techniques

It is possible to compare twO series.

3.
A
It indicates the relation between mean, mcdian and mode.

TYPES OF SKEVWNESS
533
1. Fairly symmetrical
2.
Positively skewed
3. Negatively skewed.

B34 MEASURES OF SKEWNESS

Measure of skewness is known as the measure of symmetry.
There are two types of measures of skewness.
1. Absolute measure: Absolute measure= Mean Mode -
Relative measure There are four types of relative measures of skewness.
:
2.

(i) Karl Pearson's Coefficient of Skewness

(ii) Bowley's Coefficient of Skewness.
(ii) Kelly's Coefficient of Skewness and
(iv) Measure of skewness based on the moments.
(Mode =3 Median - 2 Mean)

5.35 KARL PEARSON'S COEFFICIENT OF SKEWNESS

Mean Mode - -
Mean (3 Median 2 Mean)
Karl Pearson's Coefficient of Skewness
Standard deviation Standard deviation

3 (Mean Median)-
Standard deviation
It generally lies between -3 and 3.
If its value is zero then there is no skewness.

b36 TYPES OF SKEWNESS IN TERMS OF MEAN AND MODE

1. There is no skewness in the distribution

Mean Mode
=0 Mean - Mode = 0 Mean = Mode.
S.D.
2. The distribution is positively skewed if S, > 0.
Mean - Mode Mean - Mode > 0 Mean > Mode.
S.D.
3. The distribution is negatively skewed if
S <0
Mean Mode - Mean -
Mode <0 Mean < Mode
<0
S.D.
Xample31 Compute the coefficient of Skewnessfrom the following data:
11 12
7 8 10
X

3 13 5 4
(U.P.III Semester, 2009-2010)

2024/10/17 22.50
 Introduction to Engineering Mathematics -- V

Solution, Let a=9

d9
6 3 --3 -9 2way3 27
- 12 24 9
-2
8 18
9 13 31
10 8 8 39
11 5 2 10 20RAN 44
12 4 3 12

Efd = 0
362
124
48
f= 48 fd?=
Mean a+ Efd 9 + =9
48
Mode = ltem of maximum frequcncy (13) = 9

S.D. = ( 124
48
=1.61
r
Mean Mode
9-9 0
Karl Pearson's Coefficient of Skewness
=0 Ans.
S.D. 1.61 1.61
Example 32. Calculate Karl Pearson s Coefficient
of Skewness from the table given below:
14.5 15.5 16.5 17.5 18.5 19.5 20.5 21.5
35 40 48 100 125 87 43 229
Solution. Let a= 17.5

f d=x-17.5 fd fd? c.f.

14.5 35 -3 - 105 315 35
15.5 40 -2 -80 160
16.5 75
48 -1 - 48 48
17.5
4100 0 123
18.5 125 00
125
223
19.5 87
2 174
125348
20.5 43 348
3 435
21.5 129 387
22 478
4
88
Ef= 500 352 s00
Efd = 283 Efd'= 1735
Mean = a+ Zfd=17.5 t = 17.5 +
283
500 0.566 = 18,066
Mode = Item of
maximumfrequency=
18.5

S.D. = 1735 283 2

V
Ef V
500
500 =3.470-0.32 = 3.15=1.77
Karl Pearson's Coefficient
of Skewness Mean -Mode
18.066 -18.5
S.D.
-0.434 1.77
0.24s
0217O/17 2251 Ans.
 rStatistical Techniques
Fxample 33. Calculate Karl Pearson s Coefficient of
Skewnessfrom the table given below
Wages of day 55 - 58 58 - 61 61 -64 64-67 67 - 70
No. of workers 12 17 23 18 11

Solution. Let a=62.5

Wages of day No. of workers Mid value d=x 62.5 fd cf
(x)
55-58 12 56.5 -6 - 72 432 12
58-61 17 59.5 -3 - 51 153 29
|61-64 23 62.5 0 52
64-67 18 65.5 3 54 162 70
67- 70 11
68.5 6 66 396 81
Ef= 81 Efd=-3 fd= 1143
E
Mean = a+ fd = 62.5 + -3 -=62.5
1
= 62.46
81 27
81
2 --29 23
Median =1+ Xi= + 2 (3) = 61+
61 (3) = 62.5
23 2 (23)

2
S.D. = Efd 1143 1143 1 |10286
V
81 81 81 729
=3.76
729
Karl Pearson's Coefficient of Skewnss =
3 (Mean - Medián)
S.D.
3 (62.46--62.5) -0.12
3.76 -0.032 dtn Ans.
3.76
Example 34. Find Karl Pearson 's Coefficient of
Skewnessfor the following data:
Marks 0- 10 10 -
20 20 - 30 30-4040- 50 50 - 60
No. of students 6070
10 12 18 25 16 14 8
Solution. Let
a=35
Marks No. of
Mid value
d=x- 35 Cumulative
students ) (x) fd fa?
0-10 frequency
10 -20
10 5 - 30 - 300 9000 10
20-30
12 15 - 20 - 240 4800 22
30-40
18 25 - 10 180 1800
25
40
35 = a
40-50 0 65
16 45
S0 10
-60 160 1600 81
14 55
60-70 20 280 5600
8
65 95
30 240
2f= 103 7200 103.
24/12jd 40 fa =30,000
 Introduction to Engineering Mathematics-V
**
- 0.388 = 34.612
Mean a+ =35 + =35
103
2f103
N
-=51.5
2 2 N 40
c.f. just morc than i.c. 30
Hence, median class is class having 2
Here C= 40, /=30, f= 25, N= 103, i= 10
N_C
2
Median =l+ Xi

30 +
51.5–40 |a0) = 30 + (11.5
25

=30 + 4.6 = 34.6

2
E fd
Standard deviation =

2
30,000 (-40
V 103 103

|30,000 1600
17.062
103 10609
Karl Pearson's Coefficient of Skewness = 3 (Mean - Median) 3(34.612-34.6)
=0,0021
standard deviation 17.062
Example 35. Calculate Karl Pearson s Coefficient Ans.
of Skewness from the given data:
Life time
in months 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110| 110-120|
|No. of mobiles 4 26 28 12 5 3
Solution. Let
a=75
Life time in No. of Mid-value
months mobile (f) d=x- 75 Cumulative
() fd
frequency
30 – 40 4 35 - 40 - 160
40-50 6 6400 4
50 -60
45 - 30 - 180
8 5400 10
55 - 20 160
60 - 70 26
3200 18
65 - 10 260 2600 |44
70- 80
|28 75 0
80 -90 12 72
85 10
90- 100 8 120 84
95 20
1200
100–110 5 160
110 – 120
105
30 3200 92
3 150
115
40 4500 97
Ef= 100 120 100
4800
fd=-210
2024,A10/17 fd=31300
 Statistical Techniques
175
-210
Mean = a+ -=75 + -=72.9 l=70
100
N N= 100
100
- 44 C= 44
Median =1+ Xi =70 + 2
(10) =70 + 2.143 = 72.143
100
28 f=
i= 10
2 2
|31300 -210
S.D. = =V313-4.41
100 100
= /308.59 = 17.57
3 (Mean - Median) 3(72.9 - 72.143)
Karl Pearson's Coeficient of Skewness = S.D. 17.57
3 (0.757) 2.271
=0.1293 Ans.
17.57 17.57

EXERCISE 5.3
Calculate Karl Pearson's Coefficient of Skewness from the data given below:
1. S.D. = 6.5,AM= 29.6, mode = 27.52. Ans. S, = 0.32

Mean = 100, Variance = 35, Median = 99.61. 0.2

2 Ans. S,

3. AM = 45, Median = 48, S.D. = 22.5. Ans. S, -0.4

4. The sum of the 20 observations is 300 and sum of the squares of the observation is 5000,
Median = 15. Ans. S, =0
:

5. Find the Karl Pearson's Coefficient of Skewness for.the following

Years under 10 20 30 40 50 60
51 78 97 109
No. of persons 15 32
Ans. - 0.32
following data
6
Calculate Karl Pearson's Coefficient of Skewness from the
5.5 6.5 7.5 8.5 9.5 10.5 11.5
Cost per item (in ) 4.5
35 40 48 100 125 87 43 22
No. of items
Ans, - 0.2445

7
From the following data calculate Karl
Pearson's Coefficient of Skewness,

10 20 30 40 50 60 70 80
Scores 0
140 100 80 80 70 30 14
No. of players 150
Ans. 0.462
Skewness for the following data:
8. Find the Karl Pearson's Coefficient of
30-39 40-49 50-59 60-69 70-79 80-89
Class 10-19 20-29
14 20 25 |15 8 4
Frequency 5
at o
Ans. - 0.2064

2024/10/17 23:06
 –
Introduction to Engineering Mathematics IV

9 Lucknow students appeared in math- II and got the marks as given in the following
At
table.
:
Calculate Karl Pearson's Coefficient of Skewness from the data

Marks 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80

No. of students 10 40 20 10 40 16
14
Ans. 0.754
10. The weekly wages in of
the workers in a shoc factory are given below:

Weekly wages (in ) 500-600 600-700700-800800-900900-1000 1000-1100

No. of workers 8 12 4

Calculate Karl Pearson's Coefficient of Skewness. Ans. 0.34

11. Which of the following two series is symmetrical:
Series (a): Mean=32, Median = 34, S.D. = 20
Series (b): Mean = 32, Median = 36, S.D. = 25
Ans. series (a) is more symmetrical than serjes (h
12 Karl Pearson's Coefficient of Skewness of a distribution = 0.32, Standard deviation = 6.5
A.M, = 29.6.
From the above data find the mode and the median for the distribution.

Ans. Mode = 27.52, Median = 28.91

5.37 BOWLEYS COEFFICIENT OF SKEWNESS

Bowley's Coefficient of Skewness is
based on the quartiles and Median.
symmetrical if the distance between A distribution is
the first quartile and median is equal to
the median and third quartile i.e. the distance between

First Quartile
Median.
-2, =,- Median
Median Third Quartile

If it is not so i.e.
Median
-Q,
then the distribution
,- Median

is skewed.
5.38 MEASURE OF
BOWLEY'S COEFFICIENT
There are two ways to measure OF SKEWNESS
1. Bowey's absolute
2. Bowley's Relative
Bowley's Coefficient
measure of
Skewness =
Measure of Skewness
+
of Skewness.
-2 Median
,,
Bowley's Coefficient of Skewness =
t-2 -
Median
This formula is
also known as
Quartile Coefficient
of Skewness.
5.39 CHARACTERISTICS OF BOWLEY'S
1. If the COEFFICIENT
distribution has open OF SKEWNESS
Skewness cannot end or unequal
be calculated class intervals
2. Bowley's Coefficient and Bowley's
Coefficient of
then Pearson's Coefficient
of Skewness
lies Skewness can be calculated.
between -17a:d
-1.
 Statistical Techniques
4 Bowley's measure is calculated only from the continuous distribution with exciusive
classes.

E40 LIMITATIONS OF BOWLEY'S COEFFICIENT OF SKEWNESS

1 It is based on the central
50% of the data and ignores the renaining 50% of the data on
the extremes.
). Bowley's formulac and Pcarson's formulae cannot be
compared. However, if the distribution
is sVmmctrical then both cocfficicnts arc zero.

Bxample 36. From the following data fiund Bowley s Coefficient of Skewness:
Difference of quartiles = 80, Mode = 6)
Sum of the quartiles = 120
and Mean = 45
Solution. Here, we have
= 120
,+9,
= 80
O,-2,
Mode = 60
Mean = 45
We know that
Mode = 3
Median – 2 Mean
- 150
60 =3 Median 2
(45) Median = -= 50
3
, Median
Bowley's Coefficient of Skewness = +-2
120- 2(50) 1

80 4
= 0,25
Ans.
Erample 37. The sum of the upper quartile and lower quartile is 100 and the median is 55.
The Bowley s Coefficient of Skewness is -0.6. Find the upper and lower quartiles.

Solution. We know that

,+
Bowley's Coefficient of Skewness =
+-2 Median
. (1)
Here,
2,- 100, M= 55, S, = 0.6,
Putting the values of O, + Q,, M and Coefficient of Skewness in (l), we get

100-2 (55) - (0.6 = 100- 110

-0.6=

-10 50
- 0.6 = ...

, 3
(2)
But
+Q, = 100 .. (3)
And 50
3
On adding (2) and
(3), we get
350
20, = 3 3
Putting the value
of O, in (3), we get

2024/10/17 22:51
 Introduction to Engineering
Mathematics - |V
** t 175 125
175 100
3
+ =100 9= 33
125 175 Ans.
and
Hence, 3
3
=3 below
of Skewnessfrom the data given
Bowley s Coefficient
Example 38. Calculate
3 4 6
No. of houses 25 13 8
20 14 4
No. of Airconditioners 15

Solution.
No. of Air conditioners () Cumulative frequeney
No. of Houses (x)
15 15
0
20 35

14 49
2
3 25 74
13 87

5 8 95
6 4 99

, = size of N+1 = 25th item, Hence O, = 1

,
4

, = size of
3 (N+1) =
4
75th item. Hence =4

N+1
Median = size of =
50th item. Hence Median=3
2
Median
Bowley's Coefficient of Skewness = +-2 -

4+1-2(3) 1

4-1 3

0.33 ST2I1 Ans.

Example 39: The following
table shows the distances between
their office situated at Connaught Place, the workers residence and
New Delhi.
Distances 0- 10 10 - 20| 20 -
No. of workers 2
3030 - 40 40 - 50 S0- 6060-70|
5 10
Calculate the Bowley's 15 10 4
Coefficient of Skewness.
Solution.

Distance (x)
No. of workers
0- 10 ()
2 cf
10 -20 2
20 - 30 7

2024/10
103 2231 17
17
 Statistical Techniques T9
-- 40
30 5 32
40 -50 10 42
50 - 60) 4 46
60- 70 47
N 47 N 47 3N 141
N= 47, 2 4
4 2 4 4
The class 30 – 40 is the median class
]= 30, i= 10, f= 15, cf= 17, N = 47
N
-C
Median = 1+| 2 ()

47
-17
= 30 + 2 13 103
(10) = 30 +
15 3 3

N (47
4 99
xi = 20 + 4 (10) =20+9
10 4 4

(3N
4
-c 36-32 =
xi= 40+ a0) 40+4= 44
f 10

Bowley's Coefficient of Skewness = +-2 Median

173 99 103
4 4
--2x
3
173 99
+
4 4

(816-824)
12 8
272 816
4
=-0.0098
Ans.
Example 40 The
table shows
|Salary
in () 100-199 |200-299 300-399 400-499
No, of
workers |500-599 600-699700-799 800-899
19 25 69 62
Calculate 38 37 32
Bowley's Coefficient of Skewness.

2024/10/17 22:51
 * Introduction to Engineering
Mathematics -V
Solution.
Class boundries
No. of workers cf.
Salary in Rs. (x) 19 19
- 199 99.5- 199.5
100 25 44
199.5 - 299.5
200 - 299 69 113
299.5 - 399.5
300- 399 - 499.5 62 1.75
399.5
400 – 499 38213
– 599.5
500 - 599
499,5
37 250
- 699 599.5 - 699.5
600 32 282
- 799 699.5 – 799.5
700 18 300
- 899 799.5 – 899.5
800
N 3N =
N= =150, 225
300, 75,
Here, N= 4 2 4
599.5 – 600e
- 499.5 is the median class,
299.5 -399.5 contains 0, and
The class 399.5
contains
,
100
=399.5 + (150 – 113)
Median 62
= 399.50 + 59.68 = 459.18

(75– 44) = 299.50 + 44.93

100
,-*-c=299.5+ 69

=344.43
i3N 599.5 + 100 (225 - 213) = 599.50 + 32.43 = 631.93
,-1+-c=
fL4 37
,+-2 Median 631.93+ 344.43-2 (459.18)
Bowley's coefficient of skewness - 631.93–344.43

58 Ans.
-=0.202
287.5

EXERCISE 5.4
1. The data for a distribution is given below:
9, = 8.6, Median = 12.3, Q, 14.04
Calculate Bowley's Coefficient of Skewness. Ans, - 0.36
2. Calculate (a) Karl Pearson's Coefficient of Skewness
(b) Bowley's Coefficient of Skewness from the following
data:
City A City B Ans,
A. M.
150 140 City B
City A
Median 142 155 Karl Pearson's Coeff. -0.82
0.8 0.167
S.D. 30 55 Bowley's Coef. -0.203
195 260
62 80

2024/1017 22.51
 Statistical Techniques

Compute the quartile Coefficient of Skewness for the following dístributíon:

3.
r3-7 |8- 12 |13 17|18
-- -
22 23
-
27
28-32 133-37 |38– 42
108 580 80 32175 18 5 Ans. 0.I19
From.the following data find out Karl Pearson's Cocficicnt of Skewness and Bowley's Coefficient|
4. :
of Skewncss
S.N. Allahabad Lucknow Ans.
256.5 240.8 SN. Allahabad Lucknow
Mean
Karl Pearson's Cocff. 0.77 0.649
Median 201.0 201.6
181.1 Bowley's Coeff. 0.146 0.0385
S.D. 215.4
260.0 242.0
157.0 164.2
:
E Calculate Bowley's Cocfficient of Skewness for the following data
10 15 20 25 30 35 40 45
10 12 15 5 2 Ans. 0.33
f
:

6. blood donation camp was held at Janakpuri. The distribution of the blood donor is given below

|Age in years 20 - 25 25 30|30 - 35|35 - 40|40 – 45|45 – 50|50- 55|55 60

No. of donors 50 70 80 180 150 120 70 50

Calculate the Bowley's Coefficient of Skewness. Ans. 0.232|

a are given below :

The marks obtained by the students of public school in surprise test

a
7.

Marks 0- 10 10- 20 20- 30 30 - 40 40 - 50 50

-
6060 -70 70
-
80
10
25 20 15 10 35 25 10
No. students
Calculate the Bowley's Coefficient of Skewness. Ans. - 0.249
8
Calculate Bowley's Coefficient of Skewness for the following distribution.

Classes 1-5 6- 10 11- 15

16 - 20 21 - 25 26-30 31 - 35
3 4 68 30 10 6 2
Frequency
as first class 0.5 - 5.5]
[Hint: First change to exclusive distribution for taking real class limit
Ans. 0.262
a Government school were noted as follows:
9 In a routine check up the weights of the students of

Weights 55 - 58 58 - 61 61 –64 64-67 67 - 70

12 17 23 18 11
No. of students
Calculate the Bowley's Coefficient of Skewness. Ans. - 0.0133)
16. The participants of different ages who took part in "Marathon"
race at INDIA GATE on 2nd October
2008 as follows :

Age (in years)

- -
10
-
20| 20 30 30
-
4040 70 - 50 50 - 6060 70 80
No.of participants 358 2417 976 129 62 10

Calculate Bowley's Coefficient of Skewness. Ans. 0.131|

b41 KELLY'S CoEFFICIENT OF SKEWNESS

Bowley's measure neglects the extreme data tomeasure skewness. The entire data should be
taken
into account in measuring skewness.
Kelly modified Bowley's measure of skewness by taking any two deciles equidistant from the
median or
any two percentiles equidistant from the median.

2024/10/17 22-51
 3 Introduction to Engineering Mathematics

=
Ro +
Po -2Pso
Kelly's Coefficient of Skewness

D +D, -2Median
Kelly's Cocfficient of Skewness: D, - D

P denotes percentile and D denotes decile.

We know that, Median = Pso =Ds 18od
Generally Karl Pearson's method is widely
Note : This method is only theoretical. :
of
Skewnessfrom lhe following
ExampleA1, Calculate percentile Coefficient
P's0 = 80
Poo = 110, Po30,
Solution. Here, we have
Pso = 80
Pon = 110, Po =30,
Poo + Po -2Pso
=
Kelly's Coefficient of Skewness Poo -Po
110+30-2 (80) 140-160 -20 -1
110-30 80 80 4
= -0.25 Ans.
Example 42: Calculate Kelly 's Coefficient of
Skewnessfor the following dáta :

D = 60, D, = 290 Median = 165

Solution. Here, we have
D, = 60, D, = 290 Median= 165
Kelly's Coefficient of Skewness =
D,
+D 2Median .. (1)
D, - D

Putting the values of D,, D, and the median

in (1), we get
Kelly's Coefficient of Skewness = 60 290-2 (165)
+

290 – 60
350-330 20 2
230 23 230
= 0.087
Example 43. The weights in kg Ans.
of9 boys in a class are : 40, 42, 45. 48.
Calculate Kelly's Coefficient 50. 52, 55, 56, .
of Skewness
Solution. Here,"we have 40, based on percentiles.
42, 45, 48, 50, 52, 55,
56, 57
n=9
= Value 10
Po of (n + 1)h observation
100
= Value 10
of (9+ 1)h observation
100
= Value.
of first observation
= 40
Pco 50
S0 =Value of (n+ 1)h observation
100
= Value 50
of (9+ 1)h yervajor 2252
100
 iStatistical Techniques
= Value of 5lh observation
= 50
= Value 90
Poo of (n+ 1)th observation
100
of 90
=Value (9 + 1)h observation = Value of 9th observation
100
= 57

Now, Kelly's Coefficient

of Skewness = Po +Po -2 Pso
Poo
-Po
40 +57-2 (50)
57 – 40
97-100 3
17 17
-=-0.176 Ans.
Example 44, Calculate
the Kelly 's Coefficient of Skewnesson
the basis ofpercentiles for the
following data.
Marks obtained 20-30 30-40 40-50 50-60
No. of students 60-70 70-80 80-90 90-100
3
14 16 18 8 4
Solution.

Marks No. of students

20-30 3
cf
3
30 – 40
11
40 – 50
50 -60
20
14
60 - 70 16
34
50
70 – 80 18
68
80 -90 8
90 - 100 4
76
80
Pyo = Marks of 10 |th student= Marks of 10 80
100 th student
= Marks 100
of 8th student.
Po lies in the class 30 40 -
10
Po =+ 100
(i)=30+| 100
(10) =30+

= 30 + 50
8
=36,25
Median= Marks 80
of th student = Marks
Median Class = 21 obtained by 40th
60 -70. student
 Mathematics-IV
* Introduction to Engineering

80
N_c -34 (10) = 63.75
(10) = 60+
60+ 2
2 (E)
1+ =
P's0 16

= Marks of 90 th student
th student
Pon = Marks of

Marks of 72nd student

=
–
Po0 lies in the
class 80 90
90
(M)-C
100 (i)
Pon =/+
90
(80)- 68 72-68)
(10) = 80 + (10) = 80 + 5= 85
= 80+|J00 8

Bo + Po -2 Pso
Kelly's Coefficient of Skewness Pgo -Ro
36.25 + 85 - 2 (63.75) -6.25
85 -36.25 48.75
-0.128 Ans.

EXERCISE 5.5
Find Kelly's Coefficient of Skewness from the following table:
Poo= 200, Ps0 100 Ans. 0.143
1. Po.25,
2. D, = 15.5, D, = 120.5, Median = 70 Ans. – 0.038

3. 2.5 3.5 12.5 17.5

cf7 18 25 30
On the basis of deciles. Ans. 0.866
4. Rate (in ) Below 10 10 - 20 20 - 40 40 - 60 60 80 above 80
No. of workers 10 22 35 20 5

on the basis of deciles. Ans, - 0.07

5. Wages 800-900 | 900-1000 1000-1100 1100-1200 1300-14001400-1500
1200-1300
(in )
No. of 10 33 47 110 160 80 60
workers

on the basis of percentiles. Ans. - 0.07

6. Age (in years) below 10 10-20 20-30 30-40 40-50 50-60 .60-70 70-80 80-90
No. of patients 8 10 14 17 16 12 10 8
on the basis of percentiles. 0.07287
Ans.
Age (in years) 10- 12 12 - 14 14 - 17 20--25
17-20
No. of students 105 16
154 17
7

on the basis of deciles. Ans. 0.32

2
 as Statistical Techniques
KURTOSIS
5.42
tells about the shape of a frequency distribution. It is a measure of the flatness or peakedness
tcurve.
of the
-Leptokurtic
-
Mesokurtic

Platykurtic

The measure of kurtosis is given by B. =4 E(*-7' E(r-)

N N

5.43 TYPES OF KURTOSIS

If =3, the curve is normal or mesokurtic.
B,

B,>3, the curve is peaked than normal or leptokurtic.

B, <3, the curve is flat topped or platykurtic.

5.44 MEASURE OF SKEWNESS BASED ON MOMENTS

1. Measure of skewness is given by B, where

Karl Pearson's coefficient of skewness = t 3/2

of Karl Pearson's coefficient of skewness is determined by the sign of u,.

The sign
2. Measure of Kurtosis is given by B, where

3. Gamma Coefficients

Y,=/B,
Y, =B,-3|

arbitrary
Xample 45Find the relation between moment about the mean and moment about any
point. The first four moments of a distribution about the value 4 of the variate are
,-30 and 108. Calculate the frst four moments about the mean and find B, and B
(Uttarakhand, IlI Semester; 2008, AKTU, 20l6-17)
Solution. We
have,
= =
a4, u=-15, 17, -30 and 108
Moments about
the mean
2024,10 252
186 Introduction to Engineering Mathematics -IV

=
14.75
d (H
=
17-(-1.5)=17-2.25
+
Hy =
j-3u, 2
(u
= +
-30 -3 (17) (-1.5) 2 (-1.5)3
=- 30 + 76.5 -6.75
39.75
= - -3
Hy 4u, +6, - 3 (-1.5)*
-
108 4 (-30) (-1.5) + 6 (17) (-1.5)
108 – 180 + 229.5- 15.19 = 142.31

(4,) (39.75) 1580.06

B, = =0.4924
(H,) (14.75) 3209.05

142.31 142.31
=0.6541 Ans.
(4,) (14.75) 217.56

Example 46. Calculate the first four moments of the following distribution about the
nean
and hence fnd B, and B,:
2 3 4 7

f 28 56 70 56 28
Solution.

Calculation of Moments
d=x-4 fd fd?
-4 -4 16 - 64 256
1
8
-3 - 24 72 -216 648
2 28 -2 - 56 112 - 224 448
3 S6
4 70
-1 S6 56 - 56 56
5 0
56.
56 56
6 28 56 56
2
56 112
7
3
224 448
24 72
1
4 216 648
4 16
2f= 256 2 fd =
64 256
0 2
fd = 512 | fd = 2816
Moments about the point
x= 4 are =0 2
fd
Lfd
=)

Efd' s12
256 =2

=0

and Lfd* 2816

=1|
Moments about mean 256
are 2024 10 7 22-52
 Statistical Techniques

u,=0,

H-3u, +24= 0

-=)

11
= 2.75 Ans.
4

Eanple 41
Find out the kurtosis of the data given below:
Class-interval 0-10 10-20 20-30 30-40
Frequency 3
4 2
Solution. Let assumed mean be 25
Class Frequency Mid value
-
x,-25| 25)
fa,-25)? f-25) f-25)4
0-10 1
- 20 20 - 8000 8000 160000
10-20 3 15 10 30 -300 3000 30000
20-30 4 .25 0
30-40 35 10 20 200 2000 20000

10
Eff(,- 25)Efo-25 f(,- 25)° f,,-25)*
2f= =-30 = 900 = 9000 = 210000

Ef (x,-25) -30-=-3
10

Ef(;-25) 900
=90
10

Ef(x, - 25) -9000

-=-900
10

Ef (x -25)* 210000
= 21000
10
- = 90
–
9 = 81
Hy= '=90-(-3
= 21000 - 4(-900) (-3) + 6(90) (- 3)² -3 (-3)4
= 21000 - 10800 + 4860 - 243 = 14,817
14,817 14,817
=2.258
(81) 6561
Yo = B,--3=2.258
-3 =-0.742 Ans.

2024/ 10/A1722:52
 3 Introduction to Engineering Mathematics - IV
Example 48. Calculate the coeficient of the skewness from the following data?
Wages in Rupees 0-10 10-20 20-30 30-40 40-50 50-60 60-70
No. of labours 185 77 34 180 136 23 50
Solution. Let assumed mean be 35
Mid value
Class gelgeell
;-35 fa-35) Su-35)?
f-35
0-10 185 -30 - 5550 - 166500 - 4995000
10-20 77 15 - 20 - 1540 30800 616000
20-30 34 25 - 10 340 3400 34000
30-40 180 35 0
0
40-50 136 45 10 1360 13600 136000
50-60 23 55 20 460 9200 184000
60-70 50 65 30 1500 45000 1350000

f= 685 2f(-35) Ef,-35 2f,-35³

=–4110 = 268500 3975000
Ef(r-35) -4110
=-6
685
Ef(x-35) 268500
=391.97
685

2f(x-35) -3975000
685
-5802.9
=at u=35 –6=29
Hy
y-=391.97 - 36 =
355.97

=-5802.9 - 3 x 391.97 x
=-5802.9 + 7055.46 - 432(-6)+2(-6
=-6234.9 + 7055,46
=820.56
820,56 820.56
V
N355.97)° N45106610.72
820.56

Example 49, Calculate

given in the following
u, Hy
table and hence
Ho , 6716.15 =0.1223

for the frequencydistribution

find coeficient of skewness
of heights
Ans.

of 100 stuae
Height (cm.) and kurtosis.
144.5 149.5 - 154.5
Class interval 159,5
149:5 154.5 I59.5
164.5 - 169.5 – 174.5 -
Frequency 164.5
2 4 13
l69.5 174.5 179.5
3
32 15 3
2024/10/7 22:iB.P. II Semester 2010-2011)
 189
Statistical Techniques

Solution:
Mid x-f(x)r-)?|
Height (cmn.) f value
r(u-'
144.5- 149.5 2 147 294 --
17.2 -34.4 591.68 - 10176.89 175042.6112
149.5- 154.5 4 152 608 12.2 48.8 595.36 -7263.392 88613.3824
154.5 - 159.5 13 157 2041 --7.2 93.6 673.92 4852.224 34936.0 128
159.5 - 164.5 31 162 5022 - 2.2 - 68.2 150.04 -330.088 726.1936
164.5 - 169.5 32 167 5344 2.8 89.6 250.88 702.464 1966.899
169.5 - 174.5 15 172 2580 7.8 117 912.6 7118.28 55522.58
174.5- 179.5 3 177 531 12.8 38.4 491.52 6291.456 80530.6368
Total 100 16420 3666 -8510.39 200161.316

= 2fx l6420
=164.2
f 100

Ef (x-x) =0
100

Ef(x-I)' 3666
=36.66
f 100

Ef(r-T -8510.39
=-85.1039
100

Ef(r-)* 200161.316
100
-=2001.61316
Ef
Coefñcient -85.1039
of
theskewness, Y; =
VP3 N(36.66)
-0.3834

Kurtosis, 2001.61316 =
1.489 Ans.
(36.66)
Example 50, The first four moments of a distribution about the value 4
ure- 1.5, 17,- 30 and 108. of the variable
Find the moments about mean, B, and B,
Find also the
moments about (i) the origin, (i) the point x = 2.
SOlution, In the usual notation, we are given
assumed mean a =4
= H, =
H
=-1.5,
-
17,
H-- 30, 108

17- 2.25 = 14.75l

Hy =
=17-(-1.5)

=-30 -3 x(17) × (- 1.5) + 2(-1.5)

=-30 + 76.5 -6.75 = 39.75

= 108 - 4(- 30)(- 1.5) + 6(17) (-1.5) -3(-1.5)

-
=108 180 +
229.5 -15.1875=142.3125
2024/O/17 2253
 Introduction to Engineering Mathematics-V
190 3*
_(39.75)?
=0.4924
Hence H (14.75)

142.3125
B, = 4
=0.6541
(14.75)?
=
2.5
Also T=a+u; =4 +(-1.5)
() Moments about origin
x
We have moments about mean
X =2.5, H, = 14.75, u, =39.75 and u,= 142.31 (approx.)
u, where u is the first moment about the point =a.
x
We
know that x =at as X = mean = 2.5 =

Taking a =0, we get the first moment about origin

=X=2.5
V, = + = 14.75 + (2.5)² = 14.75 + 6.25 = 21
H
(7)
V, = Hy = 39.75 + 3(14.75)(2.5) + (2.5)³
+3u, 7 + (7)'
=39.75 + 11.0625 + 15.625 = 166
V, = t+
4u,7+6u, (7)' + (T)
= 142.3125 + 4(39.75)(2.5) + 6(14.75)(2.5)° + (2.5)
= 142.3125 + 397.5 + 553.125 + 39.0625
=1132.
(ii) Moments about the point = 2
x
We have i =a+ H. Taking a = 2, the first moment about the point x = 2 is
=
=-2= 2.5 -2 0.5
Hence, H
4t u= 14.75 + 0.25 = 15

=H+ 34, H + 3(14.75)(0.5) + (0.5)

t=39.75
=39.75 + 22.125 + 0.125 = 62

= 142.3125 + 4(39.75)(0.5) +
6(14.75)(0.5) + (0.5)
= 142.3125 + 79.5 +
22.125 + 0.0625 = 244
Example51 The first three moments about the origin are given
by
n+1 (n+)(2n +1) and w'; n(n +1)²
2
4
Examine the skewness of the data.
Solution. Here we have
n +1 +)
2s="t)2n 6
n(n +1)²
4

(n +1)(2n +1) (n+1²

6 4
2(2n* + 3n +
1)-3(1+ 2n + 1) n'-1
12 12
 Statistical Technigues

n(n +
4
l³ 3(n + 1)(2n
6
+),(0+D42"
2

n+ 2n +n 21° + 5n + 4n +1 n'+3n + 3n +1
=0
4 4 4

Coefficient
of skewness
symmetrical.
== VBn Ju =)

Ans.
The data is

Rample 52 For a distribution, the mean is 10, variance is 16, y, is 1, and B, is 4. Find

the first four'

moments about the origin.
Solution. We have,
=
7 =10, H,= 16, Y,= I, B, 4

Now, y= 1

=1>=-(16)
=
-(64)
H
64

and u, = 16
B, = 4

H=4(u,)
=
4(16) = 1024
=4
Moments about the origin

V,=10
V, =l, t x*=16+100 =116
=M, + x +
=64 + 3(16)(10) + (10 = 64 + 480 + 1000 = 1544
V, 3u,
1024 + 4(64)(10) + 6(16)(100) + (10)+
V,=4 t4u,
T +
6u, 7
+*=
= 1024 + 2560 + 9600 + 10000= 23184 Ans.
moment of a dis tribution about the value '0'
are
Example 53 The first four
mean and measure the kurtosis.
U20, 1.76, -2.36 and 10.88. Find the moments about the
mean a =0
Solution. In the usual notation, we are given assumed
and a
=0, = 0.20, u, = 1.76, u, =-2.36, u= 10.88.

Moments about mean

y =,-=1.76 - (0.20)= 1.76- 0.04

y =1.72
+ 2u
H H-3u,
=-2.36 -3 (1.76) (- 0.20) + 2 (-0.20)
–
=-2.36 + 1.056 0.016*- 1.32
192 Introduction to Engineering Mathematics-IV

2.36) (- 0.20)
+ 6(1.76)(-0.201 -3(-0.20)4
=
10.88 -4(-
0.4224-0.0048 = 9.4096
=10,88- 1.888 +
9.4096 9.4096
=3.182
Kurtosis =B, =4
; (1.72) 2.9584

Here, B, >3, so the curve is pcakcd or Leptokurtic. Ans.

EXERCISE 5.6
mean for the following data. Also calculate B, and R
1. Calculate the first four momcnts about thc
1 3 4 5 6 8 9

12 25 30 22 5 2

', - - u, =
Ans. 0.09, 2.4873, 1',= 0.018, ",= 18.212 (approx)
B, = 1.99 x 10 approx. B, = 2.9233

2. The following table gives the monthly wages of wotkers in a factory. Compute the standard deviation
and skewness.
Monthly wage (in ) No. of workers Monthly wages (in ) No. of workers
125-175 2 375-425 4
175-225 22 425-475 6

225-275 19 475-525 1

275-325 14 525-575
325-375 3

Ans. S.D =88.52, Skewness =0.7

3. Calculate the first four moments about the mean and also the value of B, from the following data.

2 3 4 5 6
8 28 156 170 56 28
Ans. u, =0, L,= 1.294, u, 0.642, M, =
0.582, B,
=395
4. Calculate the first four moments and find
the value of B, and B, from the data given below:
X 0-20 20-40 40-60 80-100
60-80
8 12 20 4

Ans. B, =0.0486, B, =2.56

5. Calculate the first four moments about
the mean and also the value of B. from the following table.
25 35 45 55 65 75 SE85
f 5 14 20 25 17 8
Ans. = 149000
540, H4
=2.3095
6. The first four moments a
B,
of distribution about the value 5 of variables are 2. 20. 40 and 50. ShoW
H,=-64, H,= 162, fB,
=1, B,= 0.63.

2024/10/1722:53
 Statistical Techniques 193
Calculate the first four moments about th0
7 mean fron the following data:
1
0
2 5 7 8
3 4 6
.10 15
15
20 25 20 10 5
valucs of B, and B, and COmment on the naturc of the distribution:
Ans. y =
2.35 B, =

B, =0, so the distribution is

=0, L, 4, ly =0, Hy=37.6, B
=0and
since symmetrical
B
<3, so the distribution is platykurtic
8 The central moments ofa distribution are Hy= , = 140,
140, L,, = 148 H, 6030. Calculate moment measures
=
of skewness and kurtosis. Ans. Y, = 0.0893, B, =0.0079
0 The first four central moments of a distribution are 0.
2.3. 0.9 and 15.65. Test the skewness and
kurtosis of the distribution.
Also discuss the nature of the curve.

Ans. Skewness = 0.265. kurtosis =2.96, The distribution is platykurtic.