28.

4 HIGHER ENGINEERING MATHEMATICS—VII

variance σ 2 then the standardized sample mean (2, 2) (2, 3) (2, 4) (2, 5)
(3, 2) (3, 3) (3, 4) (3, 5)
X−µ
Z= √ (4, 2) (4, 3) (4, 4) (4, 5)
σ/ n (5, 2) (5, 3) (5, 4) (5, 5)
is a random variable whose distribution function Now compute the statistic the arithmetic mean for
approaches that of the standard normal distribution each of these 16 samples.
N (Z; 0, 1) as n → ∞.
The set of 16 means X of these 16 samples,
gives rise to the distribution of means of the sam-
Normal Population (Small Sample) ples known as sampling distribution of means
Sampling distribution of X is normally distributed (S.D.M.):
even for small samples of size n < 30 provided sam- 2 2.5 3 3.5
pling is from normal population. 2.5 3 3.5 4
3 3.5 4 4.5
3.5 4 4.5 5
WORKED OUT EXAMPLES This S.D.M. can also be arranged in the form of
frequency distribution
Sample mean: Xi 2 2.5 3 3.5 4 4.5 5
Sampling distribution of means Frequency: fi 1 2 3 4 3 2 1
Example 1: A population consists of four num- d. The mean of these 16 means is known as mean of
bers 2, 3, 4, 5. Consider all possible distinct samples the sampling distribution of means
of size two with replacment. Find (a) the popula- 2 + 2(2.5) + 3(3) + 4(3.5) + 3(4) + 2(4.5) + 5
tion mean (b) the population standard deviation (s.d.) µ =
X 16
(c) the sampling distribution of means (d) the mean 56
of the S.D. of means (e) s.d. of S.D. of means. Verify µ = = 3.5
X 16
(c) and (e) directly from (a) and (b) by use of suitable
e. The variance of S.D.M. is
formulae.
fi (Xi − µ )2
�
Solution: σ2 = X
X n
�
a. Mean of population 1
= 1(2 − 3.5)2 + 2(2.5 − 3.5)2 +
16
2+3+4+5 14 �
µ= = = 3.5 2
4 4 · · · + 1(5 − 3.5)
b. s.d. of population
(1.5)2 + 2(1)2 + 3(.5)2 + 0+3(.5)2 + 2(1)2 + (1.5)2
� =
(2 − 3.5)2 + (3 − 3.5)2 + (4 − 3.5)2 + (5 − 3.5)2 16
σ= 10
4 = = 0.625
� 16
(1.5)2 + (0.5)2 + (0.5)2 + (1.5)2 s.d. of S.D.M. σ = 0.7905694
σ= X
4
� Verification:
5 √ i. µ = 3.5 = µ
= = 1.25 = 1.118033 X
4
ii. σ = 0.79056 = √σn = 1.118033
√
2
= 1.118033
1.4142
=
X
c. Sampling with replacement (infinite population): 0.79057
The total number of samples with replacement
is N n = 42 = 16. Here N = population size and Example 2: Solve the above Example 1 without
n = sample size. Listing all possible samples of replacement:
size 2 from population 2, 3, 4, 5 with replacement, Solution:
we get 16 samples: a. µ = 3.5
 SAMPLING DISTRIBUTION 28.5

b. σ = 1.118083 sponding means are

c. Sampling without replacement (finite popu- (2, 2) (2, 3) (2, 4) (2, 5)
lation). The total number of samples without 2 2.5 3 3.5
4!
replacement is NCn = 2!2! = 6. The six samples (3, 2) (3, 3) (3, 4) (3, 5)
are
2.5 3 3.5 4
(4, 2) (4, 3) (4, 4) (4, 5)
(2, 3) (2, 4) (2, 5) (3, 4) (3, 5) (4, 5)
3 3.5 4 4.5
(5, 2) (5, 3) (5, 4) (5, 5)
compute the statistic A.M. for each of these
3.5 4 4.5 5
samples:
compute the statistic variance for each of these 16
Xi : 2.5 3 3.5 3.5 4 4.5 samples:
The S.D.M. is variance for sample (2, 2) with mean 2 is
Xi : 2.5 3 3.5 4 4.5
fi : 1 1 2 1 1 (2 − 2)2 + (2 − 2)2
= =0
d. µ = mean of S.D.M. 2
X

2.5 + 3.0 + 2(3.5) + 4 + 4.5 21

Similarly, variance for sample (2, 3) with mean
= =
6 6
(2 − 2.5)2 + (3 − 2.5)2
= = 0.25
µ = 3.5 2
X
2 +(3−3.5)2 +2(3.5−3.5)2 +(4−3.5)2 +(4.5−3.5)2
e. σ 2 = (2.5−3.5) 6
= Thus the variance for each of the 16 samples
X
= 2.5
6
= 0.4166. 0 0.25 1 2.25
0.25 0 0.25 1
1 0.25 0 0.25
so σ = 0.645497
X 2.25 1 0.25 0
Verification:
Thus the S.D. of variances (with replacement) is

i. µ = 3.5 = µ. S2: 0 0.25 1 2.25

X Frequency: 4 6 4 2
� �
√σ N−n 1.11803 4−2
ii. σ = 0.4166 = n N−1
= √
2 4−1 Mean of S.D. of variance = 4(0)+6(0.25)+4(1)+2(2.25)
X 16
= 0.4166
10
µS 2 = = 0.625
Sampling distribution of variances 16

variance of S.D. of variance

Example 3: Find the mean and s.d. of sampling
distribution of variances (S.D.V.) for the population
4(0−0.625)2 +6(0.25−0.625)2 +4(1−0.625)2 +2(2.25−0.625)2
2, 3, 4, 5 by drawing samples of size two (a) with = 16
replacement (b) without replacement. 8.25
= = 0.515625
16
Solution: √
a. with replacement: 16 samples with their corre- s.d. of S.D. of variance = 0.515625 = 0.7180
28.6 HIGHER ENGINEERING MATHEMATICS—VII

b. without replacement: Hint: NCn = 4C2 = 6 samples are (3, 7), (3,
11), (3, 15), (7, 11), (7, 15), (11, 15).
Samples: (2, 3) (2, 4) (2, 5) (3, 4) (3, 5) (4, 5)
Means: 2.5 3 3.5 3.5 4 4.5 Ans. S.D.M.: 5, 7, 9, 9, 11, 13
Variances: 0.25 1 2.25 0.25 1 0.25 54
�
40
µ = = 9, σ = = 2.58198
Thus the S.D. of variances (without replacement) X 6 X 6

σS 2 : 0.25 1 2.25 3. Find standard error of sample means of size

Frequency: 3 2 1 2 drawn from a population 2, 3, 6, 8, 11 with
replacement.
3(0.25)+2(1)+1(2.25)
mean of S.D. of variances = 6 Hint: N n = 52 = 25 samples (2, 2), (2, 3), (2,
6), · · · · · ·, (11, 8), (11, 11).
5
µS 2 = = 0.8333 With means: 2 2.5 3 4 4.5 5 5.5 6 6.5 7 8 8.5 9.5 11
6
Frequency: 1 2 1 2 2 2 2 1 2 4 1 3 1 1
variance of S.D. of variances is
150 30
3(0.25−0.8333)2 +2(1−0.8333)2 +1(2.25−0.8333)2 µ = = 6.0, µ = = 6,
σS22 = 6
X 25 5
�
3.08333 √ 2 135
= = 0.51388 σ = 10.8 = 3.29, σ = = 2.32
6 X 25
√ Ans. S.E. = σ = 3.29
s.d. of S.D. of variance = 0.51388 = 0.71686. X
Verification:
σ2 10.8
EXERCISE µ = 6 = µ, σ 2 = 5.40 = = = 5.40
X X n 2
4. Solve the above problem without replacement.
Sampling distribution
Ans. NCn = 5C2 = 10 samples (2,3), (2, 6), (2,8),
1. Construct S.D. of means for the population 3, · · · · · ·, (6, 11), (8, 11).
7, 11, 15 by drawing samples of size two with means: 2.5, 4, 5, 6.5, 4.5, 5.5, 7, 7, 8.5, 9.5
replacement.
µ = 6 = µ, σ 2 = 4.05
Determine (a) µ (b) σ (c) S.D.M. (d) µ (e) X X
X
σ . Verify the results. σ2
� � � �
X
N −n 10.8 5−2
= = = 4.05
n N −1 2 5−1
Hint: 42 = 16 samples (3, 3), (3, 7), (3, 11),
· · · · · · , (15, 11), (15, 15) 5. Construct S.D.M. for the infinite population
with distribution given by
Ans. a. µ = 364
=9
b. σ 2
= 80
= 20, σ
= 4.4721 x: 1 2 3 4
4
c. Means: 3 5 7 9 11 13 15 f (x): 0.25 0.25 0.25 0.25
Frequency: 1 2 3 4 3 2 1 with n = 2, verify
d. µ = 14416
=9 Hint: N n = 42 = 16 samples (1, 1), (1, 2), (1,
X
2 160 3) · · · · · · (4, 3), (4, 4).
e. σ = 16 = 10, σ = 3.16227.
X X
10 5
Verification: Ans. µ = 4
= 2.5, σ 2 = 4
= 1.25
i. µ =9=µ Means: 1 1.5 2 2.5 3 3.5 4
X
ii. σ = 3.162 = √σn = 4.4721
√
2
= 3.162 Frequency: 1 2 3 4 3 2 1
X
2. Solve the above problem if sampling is without 40 10 σ2 1.25
replacement. µ = = 2.5, σ 2 = = 0.625 = =
X 16 X 16 n 2
 SAMPLING DISTRIBUTION 28.7

6. Determine the mean and s.d. of S.D. of vari- σ 0.048

σ =√ = √ = 0.008
ances for the population 3, 7, 11, 15 with X n 36
n = 2 and with sampling (a) with replacement b. without replacement
(b) without replacement
µ = µ = 22.40
Hint: N n = 42 = 16 samples (3, 3), (3, 7), X
· · · · · ·, (15, 11), (15, 15)
� �
σ N − n 0.048 1500 − 36
σ =√ = √ = 0.00790605
X n N −1 36 1500 − 1
With means: 3 5 7 9 11 13 15
Frequency: 1 2 3 4 3 2 1 σ ≈ 0.008
X
Variances: 0 4 16 36
� Example 2: Determine the expected number of
Ans. (a) µS 2 = 10, (b) σS22 = 2112
16
= 11.489 random samples having their means (a) between
7. Calculate the s.d. of S.D. of means for the 22.39 and 22.41 (b) greater than 22.42 (c) less than
population 16, 4, 12, 8, 24, 20 by drawing sam- 22.37 (d) less than 22.38 or more than 22.41, for the
ples of size 2 without replacement. Verify the above Example 1.
results.
Solution: N = size of population = 1500
Hint: S.D.M.: n = sample size = 36
Means: 6 8 10 12 14 16 18 20 22 NS = number of samples = 300
Frequency: 1 1 2 2 3 2 2 1 1 µ = population mean = 22.4,
NCn = 6C2 = 15 samples. σ = population s.d. = 0.48
Ans. µ = 84
6
= 14, µ = 210
15
= 14; use the standardized variable
X
σ = 6.8313, σ = 4.32.
X X−µ X − 22.4
8. Find mean and s.d. of S.D.M. for population 1, Z= √ =
σ/ n 0.008
2, 3 with n = 2 and sampling (a) with replace-
ment (b) without replacement. Verify results. a. For X = 22.39, Z = 22.39−22.4
= −1.26 etc.
0.0079
Ans. µ = 2, σ 2 = 23
18 1 2
a. µ = 9
= 2, σ 2 = 3
= σ2
X X
2
= 2, σ 2 = 16 = σn N−1 =
� N−n �
b. µ
X
� X� 1
= 23 · 21 · 3−2
3−1
= 6 with N = 3, n = 2

WORKED OUT EXAMPLES Fig. 28.1

Sampling distribution of means (σ known) P (22.39 < X < 22.41) = P (−1.26 < Z < 1.26)
Example 1: Determine the mean and s.d. of the = 2(0.3962) = 0.7888
sampling distribution of means of 300 random sam-
Expected number of samples = (Total number of
ples each of size n = 36 are drawn from a popula-
samples) × (probability) = (NS )P (X)
tion of N = 1500 which is normally distributed with
mean µ=22.4 and s.d. σ of 0.048, if sampling is done Expected number of samples who have mean
(a) with replacement and (b) without replacement. lying between 22.39 to 22.41 is (300)(0.7924) =
236.6 ≈ 237
Solution:
b. P (X > 22.42) = P (Z > 2.50) =
a. with replacement = 0.5 − 0.4933 = 0.0062
µ = µ = 22.40
X
28.8 HIGHER ENGINEERING MATHEMATICS—VII

Expected number of samples = (0.00057) (300) Ans. P (510 < X < 520) = P (−0.66 < Z < 1.34)
= 1.86 ≈ 2 = 0.6553
2. Calculate the probability that a random sample
of 16 computers will have an average life of
less than 775 hours assuming that length of life
of computers is approximately normally dis-
Fig. 28.2 tributed with mean 800 hours and s.d. 40 hours.
Ans. P (X < 775) = P Z < 775−800
� �
10
= −2.5
c. P (X < 22.37) = P (Z < −3.8) = = 0.0062
= 0.5 − 0.4999 = 0.0001
3. Determine the probability that X will be
between 75 and 78 if a random sample of size
100 is taken from an infinite population having
the mean µ = 76 and the variance σ 2 = 256.
75−76 78−76
Hint: Z1 = √
16/ 100
= −0.625, Z2 = √
16/ 100
Fig. 28.3 = 1.25

Expected number of samples = (300) (0.0001) Ans. P (75 < X < 78) = P (−0.625 < Z < 1.25)
= 0.03 ≈ 0 = 0.3944 + 0.2324 = 0.6268

d. P (X < 22.38) and X > 22.41) 4. Find P (X > 66.75) if a random sample of
size 36 is drawn from an infinite population
= P (Z < −2.53 and Z > 1.26) with mean µ = 63 and s.d. σ = 9.
= 0.0057 + 0.1038 = 0.1095
66.75−63
Hint: Z = √
9/ 36
= 2.5

Ans. P (X > 66.75) = P (Z > 2.5)

= 0.5 − 0.4938 = 0.0062
5. Calculate the mean and s.d. of the sampling
distribution of means of 80 samples each of
size 25 by sampling (a) with replacement
Fig. 28.4
(b) without replacement from a normal
population of 3000 with mean 68 and s.d. 3.
Expected number = (300) (0.1095) = 32.85 ≈ 33.
Ans. a. µX = 68 = µ, σ = √σ = √3 = 0.6
X n 25
�
b. µ = 68 = µ, σ = √σ N−n
EXERCISE X� X n N−1

= √3 3000−25
= 0.59759 ≈ 0.6
25 3000−1

1. Determine the probability that the sample mean 6. Determine the expected number of samples
area covered by a sample of 40 of 1 litre paint whose mean (a) lies between 66.8 and 68.3
boxes will be between 510 to 520 square feet (b) is less than 66.4 for the above Example 5
given that a 1 litre of such paint box covers Hint: z1 = (66.8−68.0) = −2.0, z2 = 68.3−68
0.6 0.6
on the average 513.3 square feet with s.d. of = 0.5
31.5 s.ft.
Ans. a. P (66.8 < X < 68.3) = P (−2.0 < Z <
Hint: Z1 = 510−513.3
√
31.5/ 40
= −0.66, Z2 = 520−513.3
√
31.5/ 40 0.5) = 0.4772 + 0.1915 = 0.6687
= 1.34
 SAMPLING DISTRIBUTION 28.9

Expected number of samples with mean standard deviation σS1 −S2 of the sampling distribu-
between 66.8 and 68.3 is = (number of tion of differences are given by
samples) (probability) = (80)(0.6680) = µS1 −S2 = µS1 − µS2 and
53.504 ≈ 54 �
σS1 −S2 = σS2 + σS2
Hint: z = 66.4−68
0.6
= −2.67. 1 2

b. P (X < 66.4) = P (z < −2.67) assuming that the samples are independent.
= 0.5 − 0.4962 = 0.0038 Sampling distribution of sum of statistics has mean
µS1 +S2 and standard deviation σS1 +S2 given by
Expected number of samples = (80)(0.0038)
= 0.304 ≈ 0. µS1 +S2 = µS1 + µS2 and
�
σS1 +S2 = σS2 + σS2
1 2
28.4 SAMPLING DISTRIBUTION
For example, for infinite population the sampling dis-
OF PROPORTIONS tribution of sums of means has mean µ and
X1 +X2
Let p be the probability of occurrence of an event σ given by
X1 +X2
(called its success) and q = 1 − p is the probability
of non-occurrence (failure). Draw all possible sam- µ =µ +µ = µ1 + µ2 and
X1 +X2 X1 X2
ples of size n from an infinite population. Compute � �
the proportion P of successes for each of these sam- σ = σ 2 + σ 2 = σ 21 + σ 22
X1 +X2 X1 X2
ples. Then the mean µP and variance σP2 of the sam- n1 n2

pling distribution of proportions are given by For sampling distribution of differences of propor-
µP = P and tions we have
pq p(1 − p) µP1 −P2 = µP1 − µP2 = p1 − p2 and
σP2 = =
n n
�
� p1 q1 p2 q2
σP1 −P2 = σP2 + σP2 = +
while population is binomially distributed, the sam- 1 2 n1 n2
pling distribution of proportion is normally dis-
tributed whenever n is large. For finite population
(with replaement) of size N , WORKED OUT EXAMPLES
µP = p and
� � Sampling distribution of differences
pq N −n
σP2 = . and sums
n N −1
Example 1: Let U1 = {2, 7, 9}, U2 = {3, 8}. Find
(a) µU1 (b) µU2 (c) µU1 +U2 (d) µU1 −U2 (e) σU1
28.5 SAMPLING DISTRIBUTION OF (f) σU2 (g) σU1 +U2 (h) σU1 −U2 . Verify that
DIFFERENCES AND SUMS µU1 + µU2 (j) µU1 −U2 = µU1 − µU2
(i) µU1 +U2 = �
Let µS1 and σS1 be the mean and standard devia- (k) σU1 ±U2 = σU21 + σU22 .
tion of a sampling distribution of statistic S1 obtained
Solution:
by computing S1 for all possible samples of size n1
drawn from population A. Similarly µS2 and σS2 be a. µU1 = 2+7+9
= 18 =6
the mean and standard deviation of sampling distri- 3 3
3+8 11
bution of statistic S2 obtained by computing S2 for b. µU2 = 2
= 2 = 5.5
all possible samples of size n2 drawn from another c. Population consisting of the sums of any member
different population B. Now compute the statistic of U1 and any member of U2 is
S1 − S2 , the difference of the statistic from all pos-
2 + 3 = 5, 7 + 3 = 10, 9 + 3 = 12
sible combinations of these samples from the two
populations A and B. Then the mean µS1 −S2 and the 2 + 8 = 10, 7 + 8 = 15, 9 + 8 = 17