Mean (X̅) = ΣX = 48 = 4.8
N 10
Standard deviation (σX) = √ΣNX2 − (ΣNX)2
= √21902 − (4180)2
= √29.2 − 23.04
= √6.16
= 2.48
C. V. = σX 100% = 2.48 100% = 51.66%
X̅ 4.8
For MBS Program
Mean (Y̅) = ΣY = 64 = 6.4
N 10
And, Standard deviation MBS(σY) = √ΣNY2 − (ΣNY)2
= √41508 − (1640)2
= √43.8 − 40.96
= √2.84
= 1.69
C. V. = σY 100% = 1.69 100% = 26.41%
̅Y 6.4
Since CV of MBS is less than CV of BBS. Thus, MBS program is easier to teach.
QN. 16) (2074) The average weekly wages, standard deviation, and number of workers of
two factories are given below.
Average weekly wage Factory A Factory B
Standard deviation Rs. 4600 Rs. 4900
No of workers Rs. 50 Rs. 40
100 80
100
Calculate the combined standard deviation. Also, explain which factory has greater
variability in the distribution of weekly wages?
Solution: Let X and Y denotes the weekly wages of factory A and factory B respectively.
Then, ̅ = 4600, ̅ = 4900, = 50, = 40, 1 = 100, 2 = 80
Now, average weekly wages of all workers
̅ 12 = ̅ 1 + ̅ 2
1 + 2
̅ 12 = 4600 × 100 + 4900 × 80
100 + 80
̅ 12 = 460000 + 392000
180
̅ 12 = 852000
180
̅ 12 = 4733.33
Again,
d1 = ̅X − ̅X12 = 4600 − 4733.33 = −133.33
and d2 = ̅Y − ̅X12 = 4900 − 4733.33 = 166.67
variance of weekly wages,
122 = 100(502 + 133.332) + 80(402 + 166.672)
100 + 80
122 = 2027688.89 + 2350311.11
100 + 80
122 = 4378000
180
σ122 = 24322.22
Coefficient of Variance of factory A
× 100% = 50 × 100% = 1.09%
̅ 4600
Coefficient of Variance of factory B
× 100% = 40 × 100% = 0.82%
̅ 4900
Since, CV(A) = 1.09% > CV(B) = 0.82, thus factory A has greater variability in wages.
QN. 17) (2073)
101
A buyer obtained samples of electronic fan from two companies A and B. He got those
samples tasted in his laboratory for length of life in the number of hours. The following are
the result of these tests.
Length of the (hours) Numbers of electronic fans
Company A Company B
600 – 800 20 6
800 1000 32 20
1000 – 1200 52 84
1200 – 1400 20 24
1400 - 1600 16 6
What would you conclude as to which supplier's fan are more uniform for length of life?
Solution: Calculation of coefficient of variance
Length of the Mid- ′ = Company A Company B
(hours) value − 1100 f2 f2d' f2d'2
(X) f1 f1d' f1d'2 6 -12 24
600 – 800 700 200 20 -40 80 20 -20 20
800 -1000 -2 32 -32 32 84 0 0
1000 – 1200 900 52 0 0 24 24 24
1200 – 1400 -1 20 20 20 6 12 24
1400 - 1600 1100 16 32 64 140 4 92
0
1300
1
1500
2
140 -20 196
Here, n1 = 140, n2 = 140, Σf1d′ = −20, Σf1d′2 = 196, Σf2d′ = 4, Σf1d′2 = 92, h = 200
Company A
̅X = A + Σf1d′ ×h= 1100 + −20 × 200 = 1100 − 28.57 = 1071.43
n1 140
σA = √Σfn1d1′2 − (Σfn11d′)2 × h
= √119406 − (−14200)2 × 200
= √1.4 − 0.02 × 200
= √1.38 × 200
= 234.95
102
( ) = × 100% = 234.95 × 100% = 21.93%
̅ 1071.34
Company B
̅X = A + Σf2d′ × h = 1100 + 4 × 200 = 1100 + 5.71 = 1105.71
n2 140
σA = √Σfn1d1′2 − (Σfn11d′)2 × h
= √19420 − (1440)2 × 200
= √0.66 − 0.0.03 × 200
= √0.63 × 200
= 158.75
( ) = × 100% = 158.75 × 100% = 14.36%
̅ 1105.71
Since, CV(A) = 21.93% > CV(B) = 14.36%
Therefore, supplier's B fan are more uniform for length of life.
QN. 18) (2073, old)
The following is the wage distribution of the workers of a company working in two
different shifts.
Morning shift Day shift
No of workers 300 200
9000
Average wage 8000 121
Variance of wages distribution 100
Find;
a) Average wage of all workers of the company.
b) Standard deviation of wages of all workers of workers of the company.
c) Coefficient of variation of all workers of the company.
Solution:
Let X and Y denotes the weekly wages of factory A and factory B respectively.
Then, ̅ = 8000, ̅ = 9000, 2 = 100, = 121, 1 = 200, 2 = 200
Now, average weekly wages of all workers
103
̅ 12 = ̅ 1 + ̅ 2
1 + 2
̅ 12 = 8000 × 3000 + 9000 × 200
3000 + 200
̅ 12 = 24000000 + 1800000
3200
̅ 12 = 25800000
3200
̅ 12 = 8062.50
Average wage of all workers of the company is 8062.50
Again,
d1 = ̅X − ̅X12 = 8000 − 8062.50 = −62.50
and d2 = ̅Y − ̅X12 = 9000 − 8062.50 = 937.50
variance of weekly wages,
122 = 1( 12 + 12) + 1( 22 + 22)
1 + 2
122 = 3000(100 + (−62.50)2) + 200(121 + 937.502)
3000 + 200
122 = 12018750 + 157781250
3200
122 = 187800000
3200
122 = 58687.50
12 = 242.26
Therefore, standard deviation of wages of all workers of workers of the company is 242.26
Finally, Coefficient of Variance of all workers of the company,
12 × 100% = 242.26 × 100% = 3.0047%
̅ 12 8062.50
QN. 19) (2072)
The mean and variance of of the marks in statistics obtained by all the 50 students of a certain
college was computed as 60 and 100 respectively. Later on, it was discovered that the score 76
was wrongly taken as 67. Find the mean and standard deviation of scores when value is omitted.
Also, calculate the coefficient of variation of marks after ignoring wrong value.
Solution: Here given
Number of items (n) = 50
mean ( ̅ ) = 60
104
variance (σ2) = 100
Incorrect values = 67
Correct values = 76
Now, we have
̅ = ∑
⇒ 60 = ∑
50
∴ = 3000
variance σ2 = ΣX2 − (ΣnX)2
n
⇒ 100 = 2 − (350000)2
50
⇒ 100 = 2 − 3600
50
⇒ 100 + 3600 = 2
50
⇒ 3700 = 2
50
⇒ 2 = 185000
Omitting the incorrect value
= 3000 − 67 = 2933
Mean = ∑X = 2933 = 59.86
n−1 49
ΣX2 = 185000 − 672 = 180511
standard deciation σ = √nΣ−X21 − (nΣ−X1)2
= √18409511 − (294393)2
= √3683.9 − 3583.22
= √100.68
= 10.03
Again, CV = σ 100% = 10.03 100% = 16.76%
̅X 59.86
105
QN. 20) (2072) Two brands of tyres are tested for their life and following result were found:
Life (000km) 20 - 24 24 - 28 28 - 32 32 - 36 36 - 40
Brand X 8 15 12 8 7
Brand Y 6 20 14 5 5
Both brands are offering same price and adverting in favor of their brand saying that the brand has
consistence of life. If you are requested to decide to purchase type of tyre of one of these two
brands, which one do you prefer and why?
Solution:
Solution: Calculation of coefficient of variance
Life (000km) Mid- ′ = Brand X Brand Y
value − 30 f1 f1d' f1d'2 f2 f2d' f2d'2
20 – 24 (X) 8 -16 32 6 -12 24
24 - 28 4 15 -15 15 20 -20 20
28 - 32 22 -2 12 0 0 14 0 0
32 - 36 888 555
36 - 40 26 -1 7 14 28 5 10 20
50 -17 69
30 0
34 1
38 2
50 -9 83
Here, n = 50, Σf1d′ = −9, Σf1d′2 = 83, Σf2d′ = −17, Σf1d′2 = 69, h = 4
For company A
X̅ = A + Σf1d′ × h = 30 + −9 × 4 = 30 − 0.72 = 29.28
n 50
σX = √Σfn1d1′2 − (Σfn11d′)2 × h
= √8530 − (−509)2 × 4
= √1.66 − 0.0324 × 4
= √1.6276 × 4
= 5.10
CV(A) = σX × 100% = 5.10 × 100% = 17.42%
X̅ 29.28
For company B
106
̅ = + 2 ′ ×ℎ = 30 + −17 × 4 = 30 − 1.36 = 28.64
2 50
σY = √Σf1d′2 − (Σfn1d′)2 × h
n
= √6509 − (−5107)2 × 4
= √1.38 − 0.12 × 4
= √1.26 × 4
= 4.49
CV(A) = σY × 100% = 4.49 × 100% = 15.68%
X̅ 28.64
Since, CV(A) = 17.42% > CV(B) = 15.68%
Therefore, I prefer to purchase tyre of brand Y, because the brand Y has consistence of life.
QN. 21) (2072 old) For the group of 200 candidates, the mean and the standard deviation were
found to be 40 and 15 respectively. Later on, it was discovered that the score 43 and 53 was
misread as 34 and 35 respectively. Find the correct mean and standard deviation.
Solution: Here given
Number of items (n) = 200
mean ( ̅ ) = 40
standard deviation ( ) = 15
Incorrect values = 34 and 53
Correct values = 43 and 35
Now, we have
̅ = ∑
40 = ∑
200
∴ = 8000
Now, Correct ΣX = 8000 − 34 − 53 + 43 + 35 = 7991
Correct mean = Correct ∑X = 7991 = 39.955
n 200
Again
107
standard deciation σ = √ΣnX2 − (ΣnX)2
r 15 = √2Σ0X02 − (8200000)2
or 225 + 1600 = ΣX2
200
or ΣX2 = 365000
correct ΣX2 = 365000 − 342 − 532 + 432 + 352
correct ΣX2 = 365000 − 342 − 532 + 432 + 352 = 364109
Correct standard deciation σ = √ΣnX2 − (ΣnX)2
= √362401009 − (7290901)2
= √1820.545 − 1596.40
= √224.145
= 14.97
QN. 22) (2071)
The combined mean and variances of salaries of 250 workers of city A and city B are 560 and
5497 respectively. The mean and variances of the salaries of 150 workers of city B are 500 and 81
respectively. Find the variance of salaries of workers of city A.
Solution:
Let X and Y denotes the salaries of city A and city B respectively.
Then ̅ 12 = 560, ( ) = 5497, ̅ = 500, ( ) = 81, = 250, 2 = 150, ( ) =?
Now, we have
̅ 1 + ̅ 2 1 = − 2 = 250 − 150 = 100,
1 + 2
̅ 12 =
560 = ̅ × 100 + 500 × 150
100 + 150
560 = 100 ̅ + 75000
250
108
140000 = 100 ̅ + 75000
100 ̅ = 140000 − 75000
̅ = 65000 = 650
100
Average salary of city A is Rs. 650.
Again,
d1 = ̅X − ̅X12 = 650 − 560 = 90
and d2 = ̅Y − ̅X12 = 500 − 560 = −60
122 = 1( 12 + 12) + 2( 22 + 22)
1 + 2
⇒ 5497 = 100( 12 + (90)2) + 150(81 + (−60)2)
100 + 150
⇒ 5497 = 100( 12 + 8100) + 552150
250
⇒ 1374250 = 100 12 + 810000 + 552150
⇒ 100 12 = 12100
⇒ 12 = 121
Therefore, variance of city A is 121.
QN. 23) (2071 old) The mode of the following distribution is Rs. 24, find the mean deviation from
mean.
Expenditure in Rs. 0 - 10 10 – 20 20 - 30 30 – 40 40 - 50
No of families 14 23 27 - 15
Solution: Let the missing frequency be a. Then
Expenditure in Rs. No of families (f)
0 – 10 14
10 – 20 23
20 – 30 27
30 – 40 a
40 – 50 15
N = 79 + a
Since, the mode of the class is 24, so the model class is 20 – 30.
Here, L = 20, f1 = 27, fo = 23, f2 = a, h = 20
109
Now, mode (Mo) = L + f1 − fo f2 × h
2f1 − fo −
⇒ 24 = 20 + 2 × 27 − 23 − × 10
27 − 23
⇒ 24 − 20 = 4 × 10
31 −
⇒ 4 = 40
31 −
⇒ 124 − 4 = 40
⇒ 124 − 40 = 4
⇒ = 21
Again, calculation of mean deviation
Expenditure No of Mid-value fX | − ̅ | | − ̅ |
families (f) (X)
in Rs. 5 70 20 280
14 15 345 10 230
0 – 10 23 25 675 0 0
10 – 20 27 35 735 10 210
20 – 30 21 45 675 20 300
30 – 40 15 =2500 | −
40 – 50 N = 100 ̅ |=1020
Now, mean(̅X) = ΣfX = 2500 = 25
n 100
Mean deviation (MD) = Σf|X − ̅X| = 1020 = 10.20
n 100
QN. 24) (2071 old) The mean weight of 150 students is 60kg. The mean weight of boys is 70 kg
with a standard deviation of 10kg. For the girls, the mean weight is 55kg with the standard
deviation 15kg. Find the number of boys and the number of girls in the class and calculate
combined standard deviation.
Solution: Let the number of boys be n1 and the number of girls be n2 respectively.
Then n1 + n2 = 150 or n1 = 150 – n2 ……….. (i)
Let the weight of boys be X1 and the weight of girls be X1.
Here, ̅X1 = 70, X̅2 = 55, σ1 = 10, σ2 = 15, ̅X12 = 60, σ12 =?
̅X12 = ̅Xn1 + ̅Yn2
n1 + n2
110
60 = 70 × (150 − 2) + 55 × 2
150
9000 = 10500 − 70 2 + 55 2
15 2 = 1500
2 = 1500 = 50
15
From equation (i)
n1 = 150 – 50 = 100
Therefore, number of boys is 100 and number of girls is 50.
Again,
1 = ̅ − ̅ 12 = 70 − 60 = 10
2 = ̅ − ̅ 12 = 55 − 60 = −5
σ122 = n1(σ12 + d12) + n2(σ22 + d22)
n1 + n2
⇒ 122 = 100(102 + (10)2) + 50(152 + (−5)2)
50 + 100
⇒ σ122 = 20000 + 12500
150
⇒ 122 = 32500
150
⇒ 122 =216.667
⇒ 12 = 14.72
Therefore, combined standard deviation is 14.72.
Analytical Question Answer
QN. 25) (2077 old)
A factory produces two types electric motors A and B. In an experiment relating to their life, the
following result were obtained:
are the result of these tests.
Life (in years) Numbers of motors
0–2 Model A Model B
2–4
4–6 15
6–8
97
12 15
11 19
111
8 - 10 89
a) Find which model of motor has greater uniformity of life? Give reason.
b) Calculate the combined mean.
c) Calculate the combined standard deviation.
Solution:
Calculation of Coefficient of variation:
Length of the Mid- ′ = − 5 Company A Company B
life (hours) value 2 f2 f2d' f2d'2
(X) f1 f1d' f1d'2 5 -10 20
0–2 -2 1 -2 4 7 -7 7
2–4 1 -1 9 -9 9 15 0 0
4–6 0 12 0 0 19 19 19
6–8 3 1 11 11 11 9 18 36
8 - 10 2 8 16 32 55 20 82
5 41 16 56
7
9
Here, n1 = 41, n2 = 55, Σf1d′ = 16, Σf1d′2 = 56, Σf2d′ = 20, Σf1d′2 = 82, h = 2
Company A
X̅ = A + Σf1d′ × h = 5 + 16 ×2 = 5 + 0.78 = 5.78
n1 41
σA = √Σfn1d1 ′2 − (Σfn11d′ 2 × h
)
= √5416 − (4116)2 × 2
= √1.3659 − 0.1521 × 2
= √1.2138 × 2
= 2.20
( ) = × 100% = 2.20 × 100% = 38.06%
̅ 5.78
Company B
112
X̅ = A + Σf2d′ ×h =5+ 20 × 2 = 5 + 0.73 = 5.73
n2 55
σB = √Σfn1d1′2 − (Σfn11d′)2 × h
= √5825 − (5250)2 × 2
= √1.50 − 0.13 × 2
= √1.37 × 2
= 2.34
( ) = × 100% = 2.34 × 100% = 40.83%
̅ 5.73
Motor of model A has greater uniformity of life, as CV(A) = 38.06% < CV(B) = 40.83%.
b) Combined mean
̅ 12 = ̅ 1 + ̅ 2
1 + 2
= 5.78 × 41 + 5.73 × 55
41 + 55
= 236.98 + 315.15
96
= 552.13
96
= 5.75
c) Combined standard deviation
1 = ̅ − ̅ 12 = 5.78 − 5.75 = 0.03
2 = ̅ − ̅ 12 = 5.73 − 5.75 = −0.02
12 = √ 1( 12 + 12) + 2( 22 + 22)
1 + 2
= √41(2.202 + 0.032 ) + 55(2.342 + (−0.02)2)
41 + 55
113
= √198.476969+ 301.18
= √4996.69569
= 2.69
QN. 26) (2077) A factory produces two types electric lamps A and B. In an experiment relating to
their life, the following result were obtained:
Length of Life (in No. of lamps A No. of lamps B
hours)
5 6
500 - 700 11 10
700 - 900 15 18
900 - 1100 10 9
1100 - 1300 8 7
1300 - 1500 6 6
1500 - 1700 5 4
1700 - 1900
Compare the variability of the length of life of two varieties of electric lamps using coefficient of
variation and state which lamp is more uniform regarding to the length of life. Also, calculate the
combined standard deviation.
Solution:
Calculation of Coefficient of variation:
Length of Mid- ′ − 1200 Company A Company B
Life (in value 200
hours) (X) = f1 f1d' f1d'2 f2 f2d' f2d'2
500 - 700 600 5 -15 45 6 - 18 54
-3 11 - 22 44 10 - 20 40
700 - 900 800 15 - 15 15 18 - 18 18
-2 10 0 0 900
900 - 1100 1000 888 777
-1 6 12 24 6 12 24
1100 - 1300 1200 5 15 45 4 12 36
0 60 -17 181
1300 - 1500 1400 60 -25 179
1500 - 1700 1600 1
1700 - 1900
1800 2
3
Here, n1 = 60, n2 = 60, Σf1d′ = −17, Σf1d′2 = 181, Σf2d′ = −25, Σf1d′2 = 179, h = 200
114
Company A
̅ = + 1 ′ × ℎ = 1200 + −17 × 200 = 1200 − 56.67 = 1143.33
1 60
σA = √Σfn1d1 ′2 − (Σfn11d′ 2 × h
)
= √16801 − (−6107)2 × 200
= √3.0167 − 0.0784 × 200
= √2.9383 × 200
= 342.83
( ) = × 100% = 342.83 × 100% = 29.99%
̅ 1143.33
Company B
̅ = + 2 ′ ×ℎ = 1200 + −25 × 200 = 1200 − 83.33 = 1116.67
2 60
= √ 1 1 ′2 − ( 11 ′)2 × ℎ
= √16709 − (−6205)2 × 200
= √2.98 − 0.18 × 200
= √2.8 × 200
= 1.67 × 200
= 334
( ) = × 100% = 334 × 100% = 29.91%
̅ 1116.67
Electric lamps B has greater uniformity of life, as CV(A) = 29.99%% < CV(B) = 29.91%.
Combined mean
̅ 12 = ̅ 1 + ̅ 2
1 + 2
115
= 1143.33 × 60 + 1116.67 × 60
60 + 60
= 68599.8 + 67000.2
120
= 135600
120
= 1130
Combined standard deviation
1 = ̅ − ̅ 12 = 1143.33 − 1130 = 13.33
2 = ̅ − ̅ 12 = 1116.67 − 1130 = −13.33
12 = √ 1( 12 + 12) + 2( 22 + 22)
1 + 2
= √60(342.832 + 13.332 ) + 60(3342 + (−13.33)2)
60 + 60
= √60(117710.09182+0 111733.689)
= √229414230.7869
= 338.71
QN. 27) (2067) Monthly income (in 000 Rs.) distribution of 400 families of a certain town
are given below:
Income, below (000 Rs.) 80 120 160 200 240 280
35 165 310 380 400
No of families 10
a) Calculate the appropriate measure of central tendency from the above distribution
and justify for your choice of measure.
b) Calculate the most suitable measures of dispersion and support for your choice.
c) Find the lowest income of richest 20% of the families.
d) Find the highest income of the poorest 20% of the families.
e) Find the limits of income of middle 50% families.
116
Solution: The given distribution is open ended class interval. The median is appropriate
measure of central tendency and quartile deviation is most appropriate measure of
dispersion.
Calculation of median and quartile deviation
Income (in 000 Rs.) No of families (f) Cumulative frequency (c.f.)
below 80 10 10
80 - 120 25 35
120 -160 130 165
160 - 200 145 310
200 – 240 70 380
240 – 280 20 400
N = 400
a) N 400
2 2
For median = = 200
cf just greater or equal to 200 is 320, so the median class is 160 – 200.
Here, L = 160, f = 145, cf = 165, h = 40
Now, median(Md) = L + N − cf × h
2 f
= 160 + 200 − 165 × 40
145
= 160 + 35 × 40
145
= 160 + 9.65517
= 169.65517
Therefore, median income is Rs169655.20
b) Calculation of quartile deviation
For lower quartile, N = 400 = 100
4 4
cf just greater or equal to 100 is 165, so the first quartile class is 120 – 160.
Here, L = 120, f = 130, cf = 35, h = 40
Now, Q1 = L + N − cf × h
4 f
117
= 120 + 100 − 35 × 40
130
= 120 + 65 × 40
130
= 120 + 20
= 140
For upper quartile, 3N = 3 × 400 = 300
4 4
cf just greater or equal to 300 is 310, so the upper quartile class is 160 – 200.
Here, L = 160, f = 145, cf = 165, h = 40
Now, upper quartile (Q3) = L + N − cf × h
4 f
= 160 + 300 − 165 × 40
145
= 160 + 135 × 40
145
= 160 + 37.2414
= 197.2414
Quartile Deviation (Q. D. ) = 3 − 1 = 197.2414 − 140 = 28.6207
2 2
Therefore, quartile deviation is Rs. 28620.70
c) Calculation of 80th percentile
80 = 80 × 400 = 320
100 100
cf just greater or equal to 320 is 380, so the median class is 200 – 240.
Here, L = 200, f = 70, cf = 310, h = 40
80 = + 80 − × ℎ
100
= 200 + 320 − 310 × 40
70
= 200 + 10 × 40
70
118
= 200 + 5.71430
= 205.71430
Therefore, lowest income of richest 20% of family is Rs. 205714.30
c) Calculation of 20th percentile
20 = 20 × 400 = 80
100 100
cf just greater or equal to 80 is 380, so the median class is 120 – 160.
Here, L = 120, f = 130, cf = 35, h = 40
20 = + 80 − × ℎ
100
= 120 + 80 − 35 × 40
130
= 120 + 45 × 40
130
= 120 + 13.8461538
= 133.84620
Therefore, highest income of poorest 20% of family is Rs. 133846.20
e) Calculation of P25 and P75
25 = 25 × 400 = 100
100 100
cf just greater or equal to 100 is 165, so the median class is 120 – 160.
Here, L = 120, f = 130, cf = 35, h = 40
20 = + 25 − × ℎ
100
= 120 + 100 − 35 × 40
130
= 120 + 65 × 40
130
= 120 + 20
= 140
119
75 = 75 × 400 = 300
100 100
cf just greater or equal to 300 is 310, so the median class is 160 – 200.
Here, L = 160, f = 145, cf = 165, h = 40
20 = + 75 − × ℎ
100
= 160 + 300 − 165 × 40
145
= 160 + 135 × 40
145
= 160 + 37.24140
= 197.24140
limits of income of middle 50% families is Rs. 140000 and Rs. 197241.40
QN. 28) (2075) What do you understand by measure of dispersion and distinguish between
absolute and relative measure of dispersion. A sample of 50 cars, each of two makes X and
Y, is taken and their average running life in years in recorded.
Life (no of years) 0 - 5 5 - 10 10 - 15 15 - 20 20 – 25
Make X 6 10 20 12 2
Make Y 8 12 17 10 3
Which of these two makes car would you prefer to buy if the costs of both makes are same
regarding consisting.
Solution:
Calculation of Coefficient of variation:
Life (in Mid- ′ = − 12.5 Make X Make Y
years) value 5
(X) f1 f1d' f1d'2 f2 f2d' f2d'2
0–5 2.5 -2 6 -12 24 8 -16 32
5 – 10 -1 10 -10 10 12 -12 12
10 - 15 7.5 0 20 0 0 17 0 0
15 - 20 1 12 12 12 10 10 10
20 - 25 12.5 2 248 3 6 12
50 -6 54 50 -12 66
17.5
22.5
Here, n1 = 50, n2 = 50, Σf1d′ = −6, Σf1d′2 = 54, Σf2d′ = −12, Σf1d′2 = 66, h = 5
120
Company A
X̅ = A + Σf1d′ ×h= 12.5 + −6 × 5 = 12.5 − 0.6 = 11.90
n1 50
σX = √Σfn1d1′2 − (Σfn11d′)2 × h
= √5540 − (−506)2 × 5
= √1.08 − 0.014 × 5
= √1.066 × 5
= 5.16
( ) = × 100% = 5.16 × 100% = 43.36%
̅ 11.9
Company B
̅ = + 2 ′ × ℎ = 12.5 + −12 × 5 = 12.5 − 1.2 = 11.30
2 50
= √ 1 1 ′2 − ( 11 ′)2 × ℎ
= √5660 − (−5102)2 × 5
= √1.32 − 0.06 × 5
= √1.26 × 5
= 1.12 × 5
= 5.62
( ) = × 100% = 5.62 × 100% = 49.73%
̅ 11.30
Electric lamps B has greater uniformity of life, as CV(X) = 43.36%% < CV(Y) = 49.73%.
QN. 29) (2074) Two companies ONIDA and BALTRA of home applications profit distribution
are given as follows:
Profit (million Rs.) 0 - 2 2 - 4 4 - 6 6 - 8 8 - 10 10 - 12
121
ONIDA 2 7 11 20 12 2
BALTRA 5 4 9 13 15 4
a) An inverter wants to invest according to the profit of the company; in which
company it is better to invest?
b) Also find the average profit and variances of both companies together.
Solution: Let the profit of ONIDA company and BALTRA company be denoted by X and Y
Respectively. Then
Calculation of Coefficient of variation:
Length of the Mid- ′ = − 5 ONIDA BALTRA
life (hours) value 2 f2 f2d' f2d'2
(X) f1 f1d' f1d'2 5 -10 20
0–2 -2 2 -4 8 4 -4 4
2–4 1 -1 7 -7 7 900
4–6 0 11 0 0 13 13 13
6–8 3 1 20 20 20 15 30 60
8 - 10 2 7 14 28 4 12 36
10 - 12 5 3 3 9 27 50 41 133
50 32 90
7
9
11
Here, n1 = 50, n2 = 50, Σf1d′ = 32, Σf1d′2 = 90, Σf2d′ = 41, Σf1d′2 = 133, h = 2
̅X = A + Σf1d′ × h = 5 + 32 ×2 = 5 + 1.28 = 6.28
n1 50
σX = √Σfn1d1′2 − (Σfn11d′)2 × h
= √9500 − (5320)2 × 2
= √1.8 − 0.41 × 2
= √1.39 × 2
= 2.36
( ) = × 100% = 2.36 × 100% = 37.58%
̅ 6.28
Again,
122
̅Y = A + Σf2d′ ×h =5+ 41 × 2 = 5 + 1.64 = 6.64
n2 50
σY = √Σfn1d1′2 − (Σfn11d′)2 × h
= √15303 − (4501)2 × 2
= √2.66 − 0.67 × 2
= √2 × 2
= 2.82
( ) = × 100% = 2.82 × 100% = 40.47%
̅ 6.64
ONIDA company is better to invest, as CV(A) = 38.06% < CV(B) = 40.83%.
b) Combined mean
̅ 12 = ̅ 1 + ̅ 2
1 + 2
= 6.28 × 50 + 6.64 × 50
50 + 50
= 314 + 332
100
= 646
100
= 6.46
Average profit of both companies is Rs. 6.64 million.
Combined variance of both companies
1 = ̅ − ̅ 12 = 6.28 − 6.46 = −0.18
2 = ̅ − ̅ 12 = 6.64 − 6.46 = 0.18
122 = 1( 12 + 12) + 2( 22 + 22)
1 + 2
= 50(2.362 + (−0.18)2 ) + 50(2.822 + 0.182)
50 + 50
123
= 280.1 + 399.24
100
= 679.34
100
= 6.7934
QN. 30) (2072 old) The following two samples describes the age of students in Model
Campus and Galaxy Campus BBS program:
Model Campus 25 31 29 24 26 23 27 28 29 26
Galaxy Campus 29 28 35 34 30 28 29 30 34 27
a. Calculate the mean and standard deviation of age of students.
b. If homogeneity in age of the students is a positive factor for learning, which of the
two campuses will be easier to teach?
c. Calculate the mean, variance, and coefficient of variation of age of students of
both campuses taken together.
d. Which campus's students are more intelligent and why?
Solution: Let the age of students' of Model college be X and that of Galaxy college be Y.
Then
Model Galaxy X2 Y2 XY
Campus (X) Campus (Y)
625 841 725
25 29 961 784 868
31 28 841 1225 1015
29 35 576 1156 816
24 34 676 900 780
26 30 529 784 644
729 841 783
23 28 784 900 840
27 29 841 1156 986
676 729 702
28 30 2 = 7238 2 =9316 =8159
29 34
26 27
= 268 =304
Model Campus
a) Mean and standard deviation of each campus
̅ = = 268 = 26.80
10
124
= √ 2 − ( )2 = √712038 − (21608)2 = √723.8 − 718.24 = 2.36
̅ = = 304 = 30.40
10
= √ 2 − ( )2 = √931106 − (31004)2 = √931.6 − 924.16 = 2.73
b) variance of each campus
( ) = 100% = 2.36 100% = 8.81%
̅ 26.80
( ) = 100% = 2.73 100% = 8.98%
̅ 30.40
Model campus will be easier to teach as, CV(X) = 8.81% < CV(Y) = 8.98%
c) Mean, variance, and coefficient of variance of both campus
̅ 12 = ̅ 1 + ̅ 2
1 + 2
= 26.80 × 10 + 30.40 × 10
10 + 10
= 268 + 304
20
= 572
20
= 28.60
Combined variance of both companies
1 = ̅ − ̅ 12 = 26.80 − 28.60 = −1.8
2 = ̅ − ̅ 12 = 30.40 − 28.60 = 1.8
122 = 1( 12 + 12) + 2( 22 + 22)
1 + 2
= 10(2.362 + (−1.8)2 ) + 10(2.732 + 1.82)
10 + 10
= 88.096 + 106.929
20
125
= 195.025
20
= 9.75125
CV of both companies
12 100% = 3.12 100% = 10.91%
̅ 12 28.60
d) Intelligence can not be determined on the basis of age of students.
QN. 31) (2072, old) The following two samples describes the age of students in morning
MBS and evening MBS program of public campus.
Morning MBS 24 30 28 23 25 22 26 27 28 25
Evening MBS 29 28 35 34 30 28 29 30 34 27
a. Calculate the mean and standard deviation of age of students of each program.
b. If homogeneity in age of the students is a positive factor for learning, which of the
two campuses will be easier to teach?
c. Calculate the mean, variance, and coefficient of variation of age of students of
both programs taken together.
d. Which programs' students are more intelligent and why?
Solution: Same as Question no 28.
126
Chapter – 5
Skewness, Kurtosis and Moments
Brief Answer Question
QN. 1) (2078) Find the Karl Pearson’s Coefficient of Skewness, when mean = 45, mode = 48 and
standard deviation = 15.
Solution: Here, mean (X̅ ) = 45, mode (Mo) = 48, and standard deviation (σ) = 15
Karl Person′s coefficient of skewness(Skp)
= 3(Mean − Median) = 3(45 − 50) = −1.5
σ 10
Since, = - 1.5 < 0, distribution is negatively skewed.
QN. 2) (2078) The quartile deviation of the distribution is 2 and the difference between P90 and P10
is 8. Calculate the coefficient of Kurtosis.
Solution: Here given Quartile deviation (Q.D.) = 2, P90 – P10 = 8
coefficient of kurtosis
K = 1 (Q3 − Q1) = 2 = 0.25
2 8
P90 − P10
QN. 2 (2077 back) If mean = 45, median = 50, and standard deviation = 10. Find the Karl Person's
coefficient of skewness and interpret the result.
Solution: Here given Mean(X̅) = 45, Median(Md) = 50 and s. d. (σ) = 10
Karl Person′s coefficient of skewness(Skp)
= 3(Mean − Median) = 3(45 − 50) = −1.5
σ 10
Since, = - 1.5 < 0, distribution is negatively skewed.
QN. 3 (2076) In a moderately skewed frequency distribution, the mean, median, and standard
deviation are 10, 8.50 and 2 respectively. Find the Karl Person's coefficient of skewness.
Solution: Here given Mean(̅X) = 10, Median(Md) = 8.50 and s. d. (σ) = 2
Karl Person′s coefficient of skewness(Skp)
= 3(Mean − Median) = 3(10 − 8.50) = 2.25
σ 2
AS, = 2.25 > 0, distribution is positively skewed.
QN. 4 (2076) First four moments about mean of a certain distribution are 0, 16, -30, and 40
respectively. Calculate the coefficient of kurtosis and interpret the result.
127
Solution: Here given μ1 = 0, μ2 = 16, μ3 = −30, μ4 = 40
Now, Coefficient of kurtosis β2 = μ4 = 40 = 40 = 0.15625
μ22 162 256
As, 2 < 3, the distribution is platykurtic.
QN. 5 (2075) For the group of 10 items ΣX = 452, ΣX2 = 24,270, and mode = 43.7. Find the
Pearson's coefficient of skewness.
Solution: Now,
Mean (X̅) = ΣX = 452 = 45.2
N 10
Standard deviation (σ) = √ΣNX2 − (ΣNX)2
= √2412070 − (41501)2 = √2427 − 2043.042 = 19.60
Pearson′s coefficient of skewness Skp = Mean − Mode = 45.2 − 43.7 = 1.5 = 0.08
SD 19.60 19.60
, = 0.08 > 0, the distribution is positively skewed.
QN. 6 (2074) First four moments about the arbitrary number 5 are – 0.1, 2.82, 0.2, and 20.58. Find
the second and third moment.
Solution: Here given μ1′ = −0.1, μ′2 = 2.82, μ3 = 0.2 and μ4 = 20.58
Now, μ2 = μ2′ − (μ1′ )2 = 2.82 – (- 0.1)2 = 2.81
μ3 = μ3′ − 3μ2′ μ1′ + 2(μ1′ )3 = 0.2 − 3 × 2.82 × −0.1 + 2(−0.1)3
= 0.2 + 0.846 + 0.002 = 1.048
QN. 7 (2073) First four moments about mean are 0, 3, 9 and 21. Test the symmetry.
Solution: Here given μ1 = 0, μ2 = 3, μ3 = 9 and μ4 = 21
Coefficient of skewness β1 = μ23 = 93 = 81 = 3
μ23 33 27
As, 1 = 3 > 0, the distribution is positively skewness.
QN. 8 (2072) List the five number summary from the following daily sales data (in Rs. 000) of
seven different shops:
Sales (in 000) 50 20 80 10 60 30 70
Shops ABCDE F G
128
Solution: Now arranging the numbers in ascending order
10, 20, 30, 50, 60, 70, 80
Numbers of shops (N) = 7
Lower quartile (Q1) = (N + 1) th item = 7 + 1 th = 2nd item = 20
4 4
Median (Md) = (N + 1) th item = 7 + 1 th = 4th item = 50
2 2
Upper quartile (Q3) = 3(N + 1) th item = 3(7 + 1) th = 6th item = 70
4 4
Five number summaries
Smallest number = 10, lower quartile = 20, median = 50,
upper quartile = 70 & largest number = 80
QN. 9 (2072) Test for the normality of the distribution on the basis of the following information
Lower quartile (Q1) = 41.5, Upper quartile (Q3) = 58.25,
10th percentile (P10) = 31.428 90th percentile (P90) = 70
Solution: Now, percentile coefficient of kurtosis
= 1 ( 3 − 1) = 1 (58.25 − 41.5) = 8.375 = 0.27
2 2 70 − 31.428 30.572
90 − 10
k = 0.27 > 0.263, the distribution is leptokurtic.
QN. 10 (2072/071)
The standard deviation of a symmetrical distribution is 7. what must be the value of the 4th
moment about the mean in order that the distribution is mesokurtic?
Solution: Here standard deviation (σ) = 7 ∴ μ2 = σ2 = 72 = 49
For the distribution to be mesokurtic
2 = 3
⇒ 4 = 3
22
⇒ 4 = 3
492
∴ 4 = 3 × 2401 = 7203
Descriptive Question Answer
129
QN. 11) (2078) Calculate the Karl Pearson’s coefficient of Skewness and interpret the result.
Class 0 - 10 10 - 20 20 - 30 30 - 40 40 - 50 50 - 60 60 – 70 70 - 80
Frequency 4 5 14 17 25 18 10 7
Solution:
Calculation of Pearsonian measure of skewness d′ = X − 35 fd' fd'2
Class Frequency c.f. Mid-value(X) 10
0 – 10 4 4 5 -3 -12 144
10 – 20 5 9 15 - 2 -10 100
20 – 30 14 23 25 -1 -14 196
30 – 40 17 40 35 0 00
40 – 50 25 65 45 1 25 625
50 – 60 18 83 55 2 36 1296
60 - 70 10 93 65 3 30 900
70 - 80 7 100 75 4 28 784
100 83 4045
Here, N = 100, A = 30, h = 10, Σfd' = 83, Σfd'2 = 4045
Now, Mean(̅X) = A + Σfd′ × h = 30 + 83 × 10 = 30 + 8.3 = 38.30
N 100
For, median N = 100 = 50
2 2
c.f. just greater or equal to 50 is 65, the median class is 40 – 50
Here, L = 40, f = 25, cf = 40, h = 10
Median (Md) = L + N − cf × h
2 f
= 40 + 50 − 40 × 10
25
= 40 + 10 × 10
25
= 40 + 4
130
= 44
Standard deviation (σ) = √Σfd′2 − (ΣNfd′ 2 × h
N
)
= √4100405 − (18030)2 × 10
= √40.45 − 0.6889 × 10
= √39.76 × 10
= 63.10
Pearsonian coefficient of skewness(Skp) = 3(X̅ − Md) = 3(38.30 − 44) = −0.27
σ 63.10
Since Skp < 0, the distribution is negatively skewed.
QN. 10 (2077) Calculate the Pearsonian measure of skewness for the following distribution:
Wage per hour (Rs.) 300-330 330-360 360-390 390-420 420-450 540-480
Number of workers 2 4 26 47 15 6
Solution:
Calculation of Pearsonian measure of skewness
Wages per No. of c.f. Mid- ′ = − 405 fd' fd'2
hour (Rs.) workers value(X) 30
300 - 330 2 2 315 -3 - 6 18
330 – 360 4 6 345 -2 - 8 16
360 – 390 26 32 375 -1 - 26 26
390 – 420 47 79 405 0 00
420 – 450 15 94 435 1 15 15
450 - 480 6 100 465 2 12 24
100 -13 99
Here, N = 100, A = 405, h = 30, Σfd' = - 13, Σfd'2 = 99
Now, Mean(X̅) = A + Σfd′ × h = 405 + −13 × 30 = 405 − 3.9 = 401.1
N 100
131
For, median N = 100 = 50
2 2
c.f. just greater or equal to 50 is 79, the median class is 390 – 420
Here, L = 390, f = 47, cf = 32, h = 30
Median (Md) = L + N − cf × h
2 f
= 390 + 50 − 32 × 30
47
= 390 + 18 × 30
47
= 390 + 11.49
= 401.49
Standard deviation (σ) = √Σfd′2 − (ΣNfd′ 2 × h
N
)
= √19090 − (−10103)2 × 30
= √0.99 − 0.0169 × 30
= √0.9731 × 30
= 29.59
Pearsonian coefficient of skewness(Skp) = 3(̅X − Md) = 3(401.1 − 401.49) = −0.04
σ 29.59
Since Skp < 0, the distribution is negatively skewed.
QN. 10 (2077, old) Calculate the Karl Pearson's coefficient of skewness for the following
distribution:
Marks 0 - 10 10 - 20 20 - 30 30 - 40 40 - 50 50 - 60 60 - 70
No. of Students 8 12 17 22 18 13 10
Solution:
Calculation of Karl Pearson's coefficient of skewness
Marks No. of c.f Mid- ′ = − 35 fd' fd'2
workers value(X) 10
132
0 – 10 8 85 - 3 - 24 72
- 2 - 24 48
10 – 20 12 20 15 - 1 - 17 17
00 0
20 – 30 17 37 25 1 18 18
2 26 54
30 – 40 22 59 35 3 30 90
299
40 – 50 18 77 45 9
50 - 60 13 90 55
60 - 70 10 100 65
100
Here, N = 100, A = 35, h = 10, Σfd' = 9, Σfd'2 = 297
Now, Mean(X̅) = A + Σfd′ × h = 35 + 9 × 10 = 35 + 0.9 = 35.9
N 100
For, median N = 100 = 50
2 2
c.f. just greater or equal to 50 is 59, the median class is 30 – 40
Here, L = 30, f = 22, cf = 37, h = 10
Median (Md) = L + N − cf × h
2 f
= 30 + 50 − 22 × 10
37
= 30 + 28 × 10
37
= 30 + 7.57
= 37.57
Standard deviation (σ) = √Σfd′2 − (ΣNfd′ 2 × h
N
)
= √210990 − (1090)2 × 10
= √2.99 − 0.0081 × 10
= √2.9819 × 10
133
= 17.27
Pearsonian coefficient of skewness(Skp) = 3(X̅ − Md) = 3(35.9 − 37.57) = −0.29
σ 17.27
Since Skp < 0, the distribution is negatively skewed.
QN. 11 (2076) The following table gives the distribution of daily wages in a company:
Wages(Rs.) 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130
No. of 10 15 20 28 35 25 18 10 8
workers
Calculate the Karl Pearson's Coefficient of Skewness. Is the distribution symmetrical? Justify.
Solution:
Calculation of Pearson's Coefficient of Skewness
Wages (Rs.) No. of c.f Mid- ′ = − 85 fd' fd'2
workers value(X) 10
40 – 50 - 40 160
50 – 60 10 10 45 -4 - 45 135
60 – 70 15 - 40 80
70 – 80 20 25 55 -3 - 28 28
80 – 90 28 0
90 – 100 35 45 65 -2 0 25
100 – 110 25 25 72
110 - 120 18 73 75 -1 36 90
120 - 130 10 30 128
8 108 85 0 32 718
169 -30
133 95 1
151 105 2
161 115 3
169 125 4
Here, N = 169, A = 85, h = 10, Σfd' = 30, Σfd'2 = 718
Now, Mean(̅X) = A + Σfd′ × h = 85 + −30 × 10 = 85 − 1.78 = 83.22
N 169
For mode
Maximum frequency is 35, thus model class is 80 - 90,
134
L = 80, f1 = 35, f0 = 28, f2 = 25, h = 10
Mode (Mo) = L + f1 − f0 f1 × h
2f1 − f0 −
= 80 + 2 × 35 − 28 25 × 10
35 − 28 −
= 80 + 7 × 10
17
= 80 + 4.12
= 84.12
Again, standard deviation(σ) = √ΣfNd′2 − (ΣNfd′)2 × h
= √171698 − (−16390)2 × 10
= √4.25 − 0.0324 × 10
= √4.22 × 10
= 20.54
Pearson′s coefficient of skewness
Skp = Mean − Mode = 83.22 − 84.12 = −0.9 = −0.044
SD 20.54 20.54
As, = −0.044 < 0, the distribution is negatively skewed. Since Skp < 0, the distribution is
negatively skewed.
QN. 12 (2075) The first four moments about the value 5 are 2, 20, 40, & 50 respectively.
Calculate mean variance, coefficient of skewness & kurtosis of distribution.
Solution: Here, A = 5, μ'1 = 2, μ'2 = 20, μ'3 = 40, & μ'4 = 50
Now, Mean (X̅ ) = A + μ'1 = 5 + 2 = 7
Variance(σ2) = μ2 = μ'2 – μ'12 = 20 – 22 = 20 – 4 = 16
First four moments about origin
μ1 = 0
μ2 = μ'2 – μ'12 = 20 – 22 = 20 – 4 = 16
μ3 = μ'3 - 3μ'2μ'1 + 2(μ'1)3 = 40 -3× 20×2 + 2 × 23 = 40 – 120 + 16 = -64
μ4 = μ'4- 4μ'3μ'1+ 6μ'2 (μ'1)2 – 3(μ'1)4 = 50 – 4×40×2 + 6 × 20 × 22 – 3 × 24
135
= 50 – 320 + 480 - 48 = 162
1 = 32 = (−64)2 =1
23 163
1 = ± √ 1 = ± √1 = ±1
As, μ3 is negative, γ1 = would also be negative.
∴ γ1 = -1 < 0, the distribution is negatively skewed.
2 = 4 = 322 = 162 = 0.63
22 162 256
Since β2 = 0.63< 3, the distribution is platykurtic.
QN. 13 (2075) The data below represent the amounts of grams of carbohydrates in a serving of
breakfast cereal. 11, 15, 23, 29, 29, 22, 21, 20, 15, 25, 17. Construct the box-and-wisher plot for
for the carbohydrate amounts.
Solution: Now arranging given number in ascending order:
11, 15, 15, 17, 20, 21, 22, 23, 25, 29, 29
Numbers of observations (N) = 11
Lower quartile (Q1) = (N + 1) th item = 11 + 1 th = 3nd item = 15
4 4
Median (Md) = (N + 1) th item = 11 + 1 th = 6th item = 21
2 2
Upper quartile (Q3) = 3(N + 1) th item = 3(11 + 1) th = 9th item = 25
4 4
Five number summaries
Smallest number = 11, lower quartile = 15, median = 21,
upper quartile = 25 & largest number = 29
136
QN. 14 (2074) Taking into account the following information, comment the nature of the
distribution:
N = 100, Σfdx = -14, Σfd2x = 154, Σfd3x = - 62, Σfd4x = 490
Solution: Now, first four moments about arbitary number are
1′ = = − 14 = − 0.14
100
2′ = 2 = 154 = 1.54
100
3′ = 3 = − 62 = − 0.62
100
4′ = 4 = 490 = 4.90
100
Again, first four moments about origin are
μ1 = 0
μ2 = μ'2 – μ'12 = 1.54 – (- 0.14)2 = 1.54 – 0.0196= 1.5204
μ3 = μ'3 - 3μ'2μ'1 + 2(μ'1)3 = - 0.62 - 3×1.54×(- 0.14) + 2 × (- 0.14)3
= - 0.62 + 0.6468 + 0.0055 = 0.0323
μ4 = μ'4- 4μ'3μ'1+ 6μ'2 (μ'1)2 – 3(μ'1)4
= 4.49 – 4×- 0.62×- 0.14 + 6 × 1.54 × (- 0.14)2 – 3 × (- 0.14)4
= 4.9-0.3472+0.18-0.001=4.73
137
Coefficient of skewness β1 = μ23 = 0.022 = 0.0004 = 0.0001
μ32 1.53 3.375
Since, β1 = 0.0001 ≈ 0, the distribution is symmetrical.
β2 = μ4 = 5.0171 = 5.2492 = 2.27
μ22 1.52042 2.3116
Since β2 = 2.27 > 3, the distribution is leptokurtic.
QN. 15 (2073) From the following distribution, find first four moments about mean, skewness,
and kurtosis of the distribution and interpret the result.
Profit in lakh (Rs.) 0 - 2 2 - 4 4 - 6 6 - 8 8 - 10
No. of companies 3 5 9 5 3
Solution:
Calculation of first four moments about mean
Profit in No. of mid- fdx fdx2 fdx3 fdx4
lakh (Rs.) companies (f) value(x)
0–2 3 13 3 33
2–4 5 3 15 45 135 405
4–6 9 5 45 225 1125 5625
6–8 5 7 35 245 1715 12005
8 - 10 3 9 27 243 2187 19683
25 125 761 5165 37721
Here, N = 25, Σfdx = 125, Σfdx2 = 761, Σfdx3 = 5165, Σfdx4 = 37721
Now, first four moments about arbitary number are
μ1′ = Σfdx = 125 = 5
N 25
μ2′ = Σfd2x = 761 = 30.44
N 25
μ3′ = Σfd3x = 5165 = 206.6
N 25
μ′4 = Σf4x = 37721 = 1508.84
N 25
Again, first four moments about origin are
138
μ1 = 0
μ2 = μ'2 – μ'12 = 30.44 – 52 = 30.44 – 25 = 5.44
μ3 = μ'3 - 3μ'2μ'1 + 2(μ'1)3 = 206.6 - 3×30.44×5 + 2 × 53
= 206.6 – 456.6 + 250 = 0
μ4 = μ'4- 4μ'3μ'1+ 6μ'2 (μ'1)2 – 3(μ'1)4
= 1508.84 – 4×206.6×5 + 6 × 30.44 × 52 – 3 × 54
= 1508.84 - 4132 + 4566 – 1875 = 67.84
1 = 32 = 0 =0
23 5.443
Since, β1 = 0, the distribution is symmetrical.
2 = 4 = 67.84 = 67.84 = 2.29
22 5.442 29.5936
Since β2 = 2.29 < 3, the distribution is platykurtic.
QN. 16 (2072) The first four moments about the value 5 are 2, 20, 40, and 50. Calculate the mean,
standard deviation, skewness, and kurtosis of distribution.
Solution: Same as QN. 12.
QN. 17 (2072) A manufacturer of battery box a sample of 13 batteries from day's production and
used then continuously until they were drawned. Numbers of hours they were used until failure
were 342, 426, 317, 545, 264, 451, 1049, 631, 512, 266, 492, 562, 298.
a) List the five number summary
b) Construct the box-and-wisher plot for data.
c) Are the data skewed? If so, how?
Solution: Now, arranging the given numbers in ascending order
264, 266, 298, 317, 342, 426, 451, 492, 512, 545, 562, 631, 1049
Number of observation (N) = 13
Lower quartile (Q1) = (N + 1) th item = 13 + 1 th = 3.25nd item
4 4
298 + 0.5(317 – 298) = 298 + 9.5 = 307.5
Median (Md) = (N + 1) th item = 13 + 1 th = 7th item = 451
2 2
Upper quartile (Q3) = 3(N + 1) th item = 3(13 + 1) th = 10.5th item =
4 4
139
545 + 0.5(562 – 545) = 545 + 8.5 = 553.5
Five number summaries
Smallest number = 264, lower quartile = 307.5, median = 451,
upper quartile = 553.5 & largest number = 1049
c) Now, Md – S = 451 – 264 = 187, L – Md = 1049 – 451 = 598
∴ Md – S < L – Md
Md – Q1 = 451 – 307.5 = 143, Q3 – Md = 553.5 – 451 = 102.5
∴ Md – Q1 > L – Md
Q1 - S = 307.5 – 264 = 43.5, L – Q3 = 1049 – 553.5 = 495.5
∴ Q1 – S < L – Q3
Since, Md – S < L – Md & Q1 – S < L – Q3, the distribution is positively skewed.
QN. 18 (2072) Find the skewness based on moments from the following data:
No. of hours worked 1-3 3-5 5-7 7-9
No. of days 35 11
Solution:
Calculation of moments:
No of No. of days Mid- fx x-x̅ Σf(x-x̅ ) f(x-x̅ )2 f(x-x̅ )3
hours (f) point
worked (x)
140
1–3 3 2 6 - 2 -6 12 -24
3–5 5 4 20 0 0 00
5–7 1 6 6 22 48
7-9 1 8 8 44 16 64
N = 10 40 0 32 48
Here, N = 10, Σf(x-x̅ ) = 0, Σf(x-x̅ )2 = 32, Σf(x-x̅ )3 = 48
Now, Mean (̅X) = Σfx = 40 = 4
N 10
μ2 = Σf(x − x̅)2 = 32 = 3.2, μ3 = Σf(x − x̅)3 = 48 = 4.8
N 10 N 10
Coefficient of skewness β1 = μ23 = 4.82 = 23.04 = 0.703125
μ32 3.23 32.768
Therefore, distribution is symmetrical.
QN. 19 (2072 old) Compute mean, variance, and skewness by using method of moments from the
following data:
Class interval 0 - 10 10 - 20 20 - 30 30 - 40 40 - 50
Frequency 4 6 10 6 4
Solution: Calculation of mean, variance, and skewness
No of No. of days Mid- fx x-x̅ f(x-x̅ ) f(x-x̅ )2 f(x-x̅ )3
hours (f) point
worked (x)
0 - 10 4 5 20 - 20 - 80 1600 - 32000
10 - 20 6 15 90 - 10 - 60 600 - 6000
20 - 30 10 25 250 0 0 00
30 - 40 6 35 210 10 60 600 6000
40 - 50 4 45 180 20 80 1600 32000
N = 30 Σfx = 750 0 4400 0
Here, N = 30, Σf(x-x̅ ) = 0, Σf(x-x̅ )2 =4400, Σf(x-x̅ )3 = 0,
Now, Mean (X̅) = Σfx = 750 = 25
N 30
μ2 = Σf(x − x̅)2 = 4400 = 146.667, μ3 = Σf(x − x̅)3 = 0 = 0
N 30 N 30
141
Variance σ2 = μ2 = 146.667
Coefficient of skewness β1 = μ32 = 0 = 0
μ32 3.23
Therefore, distribution is symmetrical.
QN. 20 (2072, old) Find the skewness and kurtosis by the methods of moments from the following
data:
weight (kg) 0 - 20 20 - 40 40 - 60 60 - 80 80 - 100
Frequencies 8 12 20 6 4
Also comment on the results obtained.
Solution: Calculation of skewness and kurtosis
Weight No. of Mid- ′ = ( − 50) fd' fd'2 fd'3 fd'4
(kg) days (f) point (X) 20 - 32 128 - 512 2048
0 - 20 8 10 -4
20 - 40 12 30 - 2 -24 48 - 96 192
40 - 60 20 50 0 0 0 00
60 - 80 6 70 2 12 24 48 96
80 - 100 4 90 4 16 64 256 1024
N = 50 -28 264 -304 3360
Here, N = 50, Σfd' = -28, Σfd'2= 264, Σfd'3 = -304, Σfd'4 = 3360
Now,
μ1′ = Σfd′ = − 28 = −0.56, μ2′ = Σfd′2 = 264 = 5.28,
N 50 N 50
μ′3 = Σfd′3 = − 304 = −6.08, μ4′ = Σfd′4 = 3360 = 67.2
N 50 N 50
μ2 = μ2′ − (μ1′ )2 = 5.28 − (−0.56)2 = 4.97
μ3 = μ3′ − 3μ2′ μ1′ + 2(μ1′ )3 = −6.08 − 3 × 5.28 × −0.56 + 2(−0.56)3
= - 6.08 + 8.87 – 0.35 = 2.44
μ4 = μ4′ − 4μ′3μ1′ + 6μ′2(μ1′ )2 − 3(μ1′ )4
= 67.2 − 4 × − 6.08 × −0.56 + 6 × 5.28 × (−0.56)2 − 3(−0.56)4
= 67.2 – 13.62 + 9.93 – 0.3 = 63.21
Coefficient of skewness β1 = μ32 = 2.442 = 5.95 = 0.0485
μ23 4.973 122.76
142
Therefore, distribution is positively symmetrical.
Coefficient of Kurtosis β2 = μ4 = 63.21 = 63.21 = 2.57 < 3
μ22 4.972 24.7
Therefore, distribution is platykurtic.
QN. 21 (2071, old) a) Pearson's coefficient of skewness for the distribution is 0.4 and its
coefficient of variance is 30%. If mode is 88, find the mean and median.
b) The first three moments of a distribution about the value 4 are 1, 4, and 10. Find the mean and
variance and, also the 3rd moment about mean.
Solution: a) Here, Coefficient of skewness (Skp) = 0.4, coefficient of variance (σ2) = 30%,
mode (Mo) = 88
Now, we have Skp = X̅ − Mo
σ
or 0.4 = ̅X − 88
σ
or σ = X̅ − 88 … … … … (i)
0.4
Again, variance(σ2) = σ × 100%
X̅
or 30% = σ × 100%
X̅
or 3̅X0==1̅Xσ03σ× 100
or
or ̅X = 3.33σ
Putting the value of σ from equation (i), we obtain
̅ = 3.33 ( ̅ 0−.488)
0.4 ̅ = 3.33 ̅ − 293.33
0.4 ̅ − 3.33 ̅ = −293.33
− 2.93 ̅ = −293.33
∴ ̅ = 100
Again, we have
143
Md = 2X̅ + Mo = 2 × 100 + 88 = 96
3 3
∴ median (Md) = 96
b) Here given A = 4, μ'1 = 1, μ'2 = 4 and μ'3 = 10
Now, Mean (X̅ ) = μ'1 + A = 4 + 1 = 5
Variance (σ2) = μ2 = μ'2 – μ'12 = 4 – 1 = 3
Third moment about mean μ3 = μ'3 – 3μ'2μ'1 + 2μ'13 = 10 – 3×4×1 + 2×1 = 10 – 12 + 2 = 0
QN. 22 (2070) If the standard deviation of a symmetrical distribution is 4. What would be the
value of fourth moment about mean in order that the distribution is a) Platykurtic b) Leptokurtic
Solution: Here standard deviation (σ) = 4 ∴ μ2 = σ2 = 42 = 16
a) For the distribution to be Platykurtic
β2 < 3
⇒ μ4 < 3
μ22
⇒ μ4 < 3
162
∴ μ4 < 3 × 256 = 768
b) For the distribution to be Leptokurtic
β2 > 3
⇒ μ4 > 3
μ22
⇒ μ4 > 3
162
∴ μ4 > 3 × 256 = 768
QN. 23 (2070) The first four moments about the arbitrary number 4 are 1, 3, 7, and 21. Find the
mean and standard deviation of the distribution.
Solution: Here given A = 4, μ'1 = 1, μ'2 = 3 and μ'3 = 7 and μ'4 = 21
Now, Mean (X̅ ) = μ'1 + A = 4 + 1 = 5
Variance (σ2) = μ2 = μ'2 – μ'12 = 3 – 1 = 2
∴ standard deviation (σ) = √2 = 1.41
Analytical Question Answer
144
QN. 24) (2078) Test the normality of these distribution of wages (in Rs.) and interpret the result
on the basis of the information given below;
Daily wages in Rs. No of workers
Factory X Factory Y
20 – 30 15 25
30 – 40 30 40
40 – 50 44 60
50 – 60 60 35
60 – 70 30 20
70 – 80 14 15
80 - 90 7 5
Solution: Calculation of Percentile coefficient of Kurtosis
Daily Factory X Factory Y cf
wages in No of workers cf No of workers
25
Rs. 65
125
20 – 30 15 15 25 160
180
30 – 40 30 45 40 195
200
40 – 50 44 89 60
50 – 60 60 149 35
60 – 70 30 179 20
70 – 80 14 193 15
80 - 90 7 190 5
200 200
For Factory X
For Q1, N = 200 = 50
4 4
cf just greater or equal to 50 is 89, so the lower quartile class is 40 – 50.
Here, l = 40, f = 44, cf = 45, i = 10
145
Q1 = l + N − cf × i = 40 + 50 − 45 × 10 = 40 + 5 × 10 = 40 + 1.14 = 41.14
4 f 44 44
For Q3, 3N = 3 × 200 = 150
4 4
cf just greater or equal to 150 is 179, so the lower quartile class is 60 – 70.
Here, l = 60, f = 30, cf = 149, i = 10
Q3 = l + 3N − cf × i = 60 + 150 − 149 × 10 = 60 + 1 × 10 = 60 + 0.33 = 60.33
4 f 30 30
For P10, 10N = 10 × 200 = 20
100 100
cf just greater or equal to 20 is 45, so the lower quartile class is 20 – 30.
Here, l = 20, f = 30, cf = 15, i = 10
P10 = l + 10N − cf × i = 20 + 20 − 15 × 10 = 20 + 5 × 10 = 20 + 1.67 = 21.67
100 30 30
f
For P90 , 90N = 90 × 200 = 180
100 100
cf just greater or equal to 180 is 180, so the lower quartile class is 60 – 70.
Here, l = 60, f = 14, cf = 179, i = 10
P90 = l + 90N − cf × i = 70 + 180 − 179 × 10 = 70 + 1 × 10 = 70 + 0.71 = 70.71
100 14 14
f
Percentile coefficient of Kurtosis
k = 1 (Q3 − Q1) = 1 (60.33 − 41.14) = 9.60 = 0.20
2 2 70.71 − 21.67 49.04
P90 − P10
Since k = 0.20 < 0.263, the distribution is platykurtic.
For Factory Y
For Q1, N = 200 = 50
4 4
cf just greater or equal to 50 is 65, so the lower quartile class is 30 – 40.
Here, l = 30, f = 40, cf = 25, i = 10
Q1 = l + N − cf × i = 30 + 50 − 25 × 10 = 30 + 25 × 10 = 30 + 6.25 = 36.26
4 f 40 40
146
For Q3, 3N = 3 × 200 = 150
4 4
cf just greater or equal to 150 is 160, so the lower quartile class is 50 – 60.
Here, l = 50, f = 35, cf = 125, i = 10
Q3 = l + 3N − cf × i = 50 + 150 − 125 × 10 = 50 + 25 × 10 = 50 + 7.14 = 57.14
4 f 35 35
For P10, 10N = 10 × 200 = 20
100 100
cf just greater or equal to 20 is 25, so the lower quartile class is 20 – 30.
Here, l = 10, f = 25, cf = 0, i = 10
P10 = l + 10N − cf × i = 20 + 10 − 0 × 10 = 20 + 10 × 10 = 20 + 4 = 24
100 25 25
f
For P90 , 90N = 90 × 200 = 180
100 100
cf just greater or equal to 180 is 180, so the lower quartile class is 60 – 70.
Here, l = 60, f = 20, cf = 160, i = 10
P90 = l + 90N − cf × i = 60 + 180 − 160 × 10 = 60 + 20 × 10 = 60 + 10 = 70
100 20 20
f
Percentile coefficient of Kurtosis
k = 1 (Q3 − Q1) = 1 (57.14 − 36.26) = 10.44 = 0.23
2 2 70 − 24 46
P90 − P10
Since k = 0.23 < 0.263, the distribution is platykurtic.
QN. 24 (2074) Calculate the skewness and kurtosis from the following wage distribution of a
factory. Explain the characteristics of the factory. Explain the frequency distribution on the basis
of skewness and kurtosis.
Solution:
Calculation of Pearson's Coefficient of Skewness
Wages per No. of Mid- ′ = − 70 fd' fd'2 fd'3 fd'4
hours (Rs.) workers value(X) 20
0 – 20 5 10 - 3 - 15 45 -135 405
20 – 40 7 30 - 2 -14 28 -56 112
147
40 – 60 16 50 - 1 - 16 16 -16 16
60 – 80 20 70 0 20 0 0 0
80 – 100 28 90 1 28 28 28 28
100 – 120 12 110 2 24 48 96 192
120 – 140 10 130 3 30 90 270 810
140 - 160 2 150 4 8 32 128 512
100
65 287 315 2075
Here, N = 100, A = 70, h = 20, Σfd' = 65, Σfd'2 = 287, Σfd'3 = 315, Σfd'4 = 2075
Now, first four raw moments are
μ1′ = Σfd′ × h = 65 × 20 = 13
N 100
μ2′ = Σfd′2 × h2 = 287 × 202 = 1148
N 100
μ3′ = Σfd′3 × h3 = 315 × 203 = 25200
N 100
μ4′ = Σfd′4 × h4 = 2075 × 204 = 3291200
N 100
Again, first four central moments are
1 = 0
μ2 = μ'2 – μ'12 = 57.4 – 132 = 1148 - 169 = 979
μ3 = μ'3 - 3μ'2μ'1 + 2(μ'1)3 = 25200 - 3×1148×13 + 2 × 133
= 25200 – 44772 + 4394 = -15178
μ4 = μ'4- 4μ'3μ'1+ 6μ'2 (μ'1)2 – 3(μ'1)4
= 3291200 – 4×25200×13 + 6 × 1148 × 132 – 3 × 134
= 3291200 - 1310400 + 1164072 – 85683 = 3059189
1 = 32 = (−15178)2 = 230371684 = 0.2455
23 9793 938313739
γ1 = ± √ 0.2455 = ±0.5
Since, μ3 is negative, γ1 would be negative, the distribution is negatively skewed.
148
2 = 4 = 3059189 = 3059189 = 3.19
22 9792 958441
Since β2 = 3.19 > 3, the distribution is leptokurtic.
QN. 25 (2073) The manager of Bakery Café selected a random sample of 50 customers' waiting
time is recorded as follows:
29, 28, 51, 43, 24, 40, 52, 72, 41, 23, 25, 30,
22, 34, 19, 31, 32, 29, 45, 24, 60, 48, 19, 47,
54, 68, 17, 43, 23, 56, 39, 40, 43, 48, 56, 42,
21, 36, 24, 65, 60, 31, 50, 31, 47, 43, 30, 32,
35, 39
a) Develop a frequency distribution using 7 classes.
b) Comment on the nature of the distribution of customers' waiting in the café.
Solution: Frequency distribution table
Waiting time (in hours) Tally marks Frequency
10 – 20 ||| 3
20 – 30 |||| |||| | 11
30 – 40 |||| |||| || 12
40 – 50 |||| |||| ||| 13
50 - 60 |||| | 6
60 - 70 |||| 4
70 - 80 | 1
Calculation of skewness and kurtosis
Waiting Frequency Mid- ′ ( − 45) Σfd' Σfd'2 Σfd'3 Σfd'4
time (in (f) value(X) 10
hours) = -9 27 -81 243
-22 44 -88 176
10 – 20 3 15 - 3 -12 12 -12 12
0 000
20 – 30 11 25 - 2 6 666
30 – 40 12 35 - 1
40 – 50 13 45 0
50 - 60 6 55 1
149