Dear Students ,

Please follwo the link but next week onward as per your weekly timetable you join session using
blackboard. I'll not be sending any link. Attendance is mandatory. While preparing your debar list for
end term these attendance would be considered.

https://au.bbcollab.com/guest/6da93db15fa04c1c95bd5ea6e6652832

One Tailed test concerning Single Mean

Q. The weight of costly catalyst electrodes purchased by a

chemical company follows normal distribution. The sales
manager of the vendor firm claims that the mean weight of
electrodes is 150 gm. The quality manager of the chemical
company wants to verify this claim. So, he has taken a sample
of 25 electrodes. The mean and the variance of the
electrodes in the sample are found to be 145 gm and 100 gm,
respectively. Verify the claim of the sales manager at a
significance level of 0.05.
Sol.
Given
Sample Size, n =25
Population Mean, µ =150gm
Sample Mean, X́ =145gm
Variance of the Sample, S2 =100gm
Standard Deviation of the Sample, S =10gm
Significance Level, α =0.05
So, X ~ N (µ, σ2). Since the sample size, n is less than 30, X́ ~ t distribution with (n-1)
degrees of freedom whose statistic is given below:
X́ −μ
t= with 24 degrees of freedom.
S /√ n

Now, null and alternate hypotheses are:

H0: µ = 150
H1: µ ≠ 150

X́ −μ 145−150
t= = = -2.5
S /√ n 10/√ 25

The table t values with the degrees of freedom of 24 at α/2= 0.025 placed at left and
right tails are -2.064 & +2.064, respectively, which are shown in fig. along with the
acceptance and rejection regions.
Since the calculated t values (-2.5) is less than -tα/2 (-2.064), the calculated t value falls in
the rejection region. Hence, H0 should be rejected.
This means that the mean weight of the catalyst electrodes is significantly different from
150gm which means that the claim of the sales manager of the vendor firm is not true.
One tailed test concerning Difference between two means
8.16- The personnel manager of a firm is recruiting its graduate trainees from colleges
affiliated to two different universities. The performance index (0-10 scale) of the
trainees from each university follows normal distribution. The variance of the
performance index of the trainees from the University-1 is 9 and that of the trainees
from the University-2 is 4. The manager feels that the mean performance index of the
trainees from the University-1 is less than that of the trainees from the University-2. To
test his intuition, he has selected a sample of 49 trainees from the University-1 and their
mean performance index is found to be 7. Similarly, he has selected a sample of 64
trainees from the University-2 and their mean performance index is found to be 6. Test
the intuition of the manager at a significance level of 0.05.
Soln.
We have
Variance of the performance index of the trainees from the University-1,σ 12 =9
Variance of the performance index of the trainees from the University-2,σ 22 =4
Number of trainees in the sample from University-1, n1 =49
Number of trainees in the sample from University-2, n2 =64
Mean performance index of the sample-1, X 1 =7
Mean performance index of the sample-2, X 2 =6
Significance level, α =0.05
We also have X1 N( μ1 , σ 21) and X2 N( μ2,σ 22).
Therefore,
σ 21 σ 22
X1 - X2
(
N μ1−μ 2 , +
n1 n2 )
σ 12 σ 22 9 4
σ 2X 1−X 2= + = 49 + 64 = 0.246
n1 n2
 2
√σ X 1−X 2 =0.496

Null and alternate hypotheses are:

H0: μ1 ≤ μ2 or μ1- μ2≤ 0
H 1: μ1> μ2 or μ1 - μ2>0

Z X 1−X 2=¿ ¿ = 2.02

The standard normal statistic for the given significance level, (α) of 0.05, Z α is 1.64,
which is shown in Fig.

Since the calculated Z value (2.02) is more than Z α(1.64), the calculated Z value falls in
the rejection region. Hence, H 0 should be rejected this means that the performance
index of the trainees from the University-1 is not less than that of the trainees from the
University-2.
Two tailed test concerning Difference between two means(when variance of the
populations are known and the sample size are large)
Example 8.18 An automobile company presently buys piston from vendor-1 .But the
quality manager is desiring to consider the supply of piston from vendor-2 to meet the
extra demand, provided the new vendor can supply the piston with good precision on
its axial length on par with that of vendor-1.If the axial length is more than the
necessity. It will lead to knocking at the top of the cylinder head. Hence it should be
avoided. The axial length of the piston supplied by Vendor-1 as well as Vendor-2 follow
normal distribution with variances of 5mm and 9mm respectively.
He has selected 64 pistons from vendor-1 and found that the mean axial length of the
sample as 60 mm. Similarly he selected 81 piston from Vendor-2 and found that the
mean axial length of this sample as 61 mm. Check whether the quality manager can
consider the supply from the Vendor-2 at a significance level 0.01

Solution:
We have
Variance of axial length of piston from Vendor-1,σ 1 x2 =5mm
Variance of axial length of piston from Vendor-2,σ 2 x 2 =9mm
Sample size of Vendor-1,n1 =64
Sample size of vendor-2,n2=81
Mean axial length of sample-1, X 1 =60mm
Mean axial length of sample-2, X 2=61
Significance level ,α=0.01
Where X1 ̴N(µ1,σ12) and X2 ̴N(µ2,σ22).Therefore,
σ 12 σ 22
X1-X2 ̴N (µ1-µ2, + )
n1 n2

σ 12 σ 22 5 9
σ X1-X22 = + = 64 + 81 = 0.1892
n1 n2

σ X1-X2 = 0.435
Now the null and alternate hypothesis are:
 H0 : µ1 ≤ µ2 or µ1 - µ2 ≤ 0
H1 : µ1 > µ2 or µ1 - µ2 > 0

( X 1−X 2 )−(µ 1−µ 2) ( 60−61 ) −0

Z X1-X2 = = = -2.3
σ X 1−X 2 0.435

Since the calculated Z value(-2.3) is in between -2.57 and +2.57, the calculated Z value
falls in the acceptance region .Hence Ho should be accepted.

This means that the mean axial length of the piston supplied by Vendor-2 does not differ
from 60mm.So, the quality manager can consider the supply from vendor-2 also.
One –tailed tests Concerning Differences between two means (when the variances of
the populations are unknown and the sample sizes are large)
8.19 The production rate of shafts produced in machine-A as well as in Machine-B
follows normal distribution. The variance of the production rate of the shafts produced
using each machine is not known. The production manager feels that the performance
of the machine-A is not comparable with that of the Machine-B. To test his intuition, the
quality assistant of the shop has selected 81 days’ record of the machine-A and found
that the mean and variance of the production rate of this sample are 70 units and 9
units, respectively. Similarly, he has selected 64 days’ records of the machine-B and
found that the mean and variance of the production rate of this sample are 68 units and
25 units, respectively. Check the intuition of the production manager at a significance
level of 0.05.
Solution.
Sample size on machine-A, n1 = 81 days’ records
Sample size on machine-B, n2 = 64 days’ records
Mean production rate of the sample from Machine-A, X1 = 70 units
Variance of the production rate of the sample from Machine-A, S 12 = 9 units
Mean production rate of the sample from Machine-B, X2 = 68 units
Variation of the production rates of the sample from Machine-B, S 22 = 25 units
Significance level, α =0.05
Where X1 ̴N(µ1,σ12) and X2 ̴N(µ2,σ22).Therefore,
σ 12 σ 22
X1-X2 ̴N (µ1-µ2, + )
n1 n2

Since the sample sizes, n1 and n2 are more than 30, the respective variances of the
samples can be substituted for the respective variances of the populations. Then
σ 12 σ 22 S 12 S 22 9 25
σ X1-X22 = + = + = 81 + 64 = 0.5017
n1 n2 n1 n2

or
σ X1-X2 = 0.7083
Now the null and alternate hypothesis are:
H0 : µ1 ≤ µ2 or µ1 - µ2 ≤ 0
H1 : µ1 > µ2 or µ1 - µ2 > 0
Hence
( X 1−X 2 )−(µ 1−µ 2) ( 70−68 ) −0
Z X1-X2 = = = 2.824
σ X 1−X 2 0.7083
The table standard normal value, Zα is 1.64, which is shown below

Two tailed test concerning difference between two means (when variance of the
populations are unknown and the sample size are large)
8.20- The regional manager of a bank wants to analyze the number of delayed payments
of instalments of loans by farmers which were availed by them under a special loan
scheme in two of its branch banks (branch-1 and branch-2) in rural areas. The number of
delayed payments of instalments in each branch bank follows normal distribution. The
manager fels that the number od delayed payments of installments by the customers of
the Branch-1 is no way different from that of the Branch-2 . So, he selected the loan
accounts of 50 different customers from the Branch-1 andfound that the mean of the
number of selected the loan accounts of 80 different customers from the Branch-2 and
found that the mean of the number of delayed payments of installments and its
varience were 23 and 36, respectively. Test his intuition at a significance level of 0.01.

Soln. given that

Number of customers selected from branch-1, n 1 = 50
Number of customers records selected from Branch-2, n2 = 80
Mean of number of delayed payments of installments in Branch-1, X́ 1 = 21
Mean of number of delayed payments of installments in branch-2, X́ 2 = 23
Variance of mean number of delayed payments of installments in Branch -1, S 12 =9
Variance of mean of number of delayed payments of installments in Branch-2, S 22 =36
Significance level, α =0.01
For expected value of difference in sample means = μ1- μ2,
2 σ 21 σ 22
σ X 1− X 2 = +
n1 n 2

If X1 N ( μ1,σ 12) and X2 N ( μ2, σ 22)

Then,
σ 2 σ 22
X1 – X2 N ( μ1- μ2, 1 + ¿
n 1 n2

Since the sample sizes ,n1 and n2 are larger, their variances S21 and S22 can be substituted
for σ 12∧σ 22. Then
2 σ 21 σ 22 S 21 s22 9 36
σ X −X = + = + = + = 0.63
1 2
n1 n 2 n1 n2 50 80

σ .X − X = 0.794
1 2

Their null and alternate hypotheses are:

H0: μ1 ≤ μ2 or μ1- μ2≤ 0
H 1: μ1> μ2 or μ1 - μ2>0

( 21−23 ) −0
Z X 1−X 2=¿ ¿ = =¿-2.52
0.794
a a
The table standard normal values , -Z 2 and Z 2 are -2.57 and +2.57.

Since the calculated Z value (-2.52) is in between -2.57 and +2.57, the calculated Z value
falls in the acceptance region .Hence H0 should be accepted.
This means that the number of delayed payments of installments by the customers of
the Branch-1 is no way different from that of the Branch-2.Hence the regional managers
intuition is correct.
Two tailed test concerning difference between two means(when the variance of the
populations are unknown and the sample sizes are small)
8.22- The diameter of a particular type of bearing supplied by vendor -1 and vendor-2
follows normal distribution .The quality manager of the firm buying the bearings
believes that the diameter of the bearings supplied by the vendor-1 is not different from
that of the bearings supplied by the vendor-2. So, the quality assistant under him
selected 16 bearings from supply of the vendor-1 and found that mean and variance of
the diameter were 38.5mm and 2.5mm respectively. Similarly, he selected 20 bearings
from the supply of the vendor-2 and found that mean and variance of the diameter
were 40mm and 4mm. Verify the intuition of the manager at a significance level of 0.10.
Soln. we have
Sample size of Vendor-1,n1 =16
Sample size of Vendor-2,n2 =20
Mean diameter of the sample from vendor-1, X 1 =38.5mm
Variance of the mean diameter of the sample form Vendor-1, S21
=2.5mm
Mean diameter of the sample form vendor-2, S22
=40mm
Variance of the mean diameter of the sample from vendor-2, S22
=4mm
Significance level, α =0.10

2 ( n1−1 ) s21 +(n2−1)s22 ( 16−1 ) 2.5+ ( 20−1 ) 4

S=p = = 3.338
n1+n −2 2
16 +20−2

s p= 1.827

The standard deviation for the difference between the means of the populations is
computed as shown below:
1 1 1 1
=Sp
√ √ n1
+
n2
= 1.827
16
+
20 √ √
=0.6128

The null and alternate hypotheses are:n

H 0 : μ1= μ2 or μ1- μ2= 0

H 1: μ1≠ μ2 or μ1- μ2≠ 0

Then t distribution is
( X 1− X 2 )−(μ1−μ2 )
t=
σ X −X
1 2

=
( X 1− X 2 )−(μ1−μ2 )
S p X √( 1/n 1) + √ ( 1 /n 2 )

= ( 38.5−40 )−0
=-2.45
0.6128

The table value of -t a and t a with

2 2
34 degrees of freedom are -1.645 and +1.645 which are as shown
in fig. along with rejection and acceptance region.
The calculated t value (-2.45) is not in between -t a2 and t a2 (-1.645 and +1.645)

Hence reject H 0.
This means that diameter of the bearings supplied by the vendor-1 is different from that
of the bearings supplied by the vendor-2.So, the quality manager’s intuition is not
correct.

Two tailed test concerning Single Proportion

8.25. The sales representative of pharmaceutical company claims that a new drug cures
fever in 90% of the cases. When it is administered by a doctor to 80 patients, it cured
only 68 patients. Check the claim of the sales representatives at a significance level at
0.05.
Sol:
We have sample size, n=80; proportion, p=0.90; and sample proportion, p, which
follows normal proportion, p=68/80=0.85. The sample proportion, p, which follows
normal distribution is as shown below, because np as well as n(1-p) is more than 5:
p (1− p)
[
p ~ N p,
n ]
 p(1−p)
σ p=
√ n

0.9(1−0.9)
=
√ 80
= 0.03354

When the significance level, α= 0.05, null and alternate hypotheses are:
H0: p=0.9
H1: p≠0.9
The standard normal statistic for p is as shown below:
p− p
Zp=
σp

0.85−0.90
= 0.03354 =-1.49

The standard normal statistics for -Zα/2 and + Zα/2 are -1.96 and +1.96, respectively which
are shown in fig:
Since the calculated Z value (-1.49) falls in the acceptance region, H0 should be accepted.
This means that the sales representative’s claim is true.
One tailed tests concerning difference between two proportions
8.26. The number of defective pieces of a bought-out item supplied by Vendor-1 and
Vendor- 2 follows binomial distribution. The quality manager of the firm buying the item
believes that the proportion of the number of defective pieces in the lot supplied by the
Vendor-1 is less than the proportion of the number of defective pieces in the lot
supplied by the Vendor-2. So, the quality assistant under him selected a sample of 120
pieces from the supply of the Vendor-1 and found that the number of defective pieces
in it is 15. Similarly, he selected a sample of 100 pieces from the supply of the Vendor-2
and found that the number of defective pieces is 11. Check the intuition of the quality
manager at the significance level of 0.05.
Sol.
Given That:
Size of sample of the Vendor-1, n1 =120
Size of sample of the Vendor-2, n2 =100
Number of Defective pieces in the Sample-1 =15
Number of Defective pieces in the Sample-2 =11
Significance Level, α =0.05
Let p1 , p2 be the proportion with respect to Vendor-1 and Vendor-2, respectively. Then
15 11
p1 = =0.125 , p2= =0.11
120 100

p1 (1−p 1)
[
p1 N p1 ,
n1 ]
p2 (1−p 2)
[
p2 N p 2 ,
n2 ]
p1 (1− p 1) p2 (1− p2 )
[
p1− p2 N ( p1− p2 ) ,
n1
+
n2 ]
 p1 (1− p1 ) p2 (1− p2 )
σ p −p =
1 2
√ n1
+
n2

0.125 X 0.875 0.11 X 0.89

=
√ 120
+
100
=0.0435

Null & alternate hypotheses

H0:p1-p2≤0
H1: p1-p2›0
The standard normal statistics for (p1-p2 ) is as shown below:
( p1 −p 2 )−( p 1− p2) ( 0.125−0.11 )−0
Z p −p = = =0.345
1 2
σp 1−p 2
0.0435

The standard normal statistics for +Zα is 1.64, which is shown:

Since the calculated Z value (0.345) is less than Zα(1.64), it falls in the acceptance region.
Hence, H0 should be accepted.
8.30 The weight of cement bags produced by Vignesh Cement Company follows normal
distribution. The quality assistant at the final inspection section of the company feels
that the variance of the weight of the cement bags has increased from a specified
maximum variance of 0.64 kg which will lead to customer complaints. Hence, he has
selected a sample of 8 cement bags and found that the variance of the sample is 0.36
kg. Check the intuition of the quality assistant at a significant level of 0.01.
Solution. Weight of cement bags follows normal distribution. We also have variance of
the weight of cement bags of the population, σ2 = 0.64 kg; sample size, n=8; variance of
the weight of cement bags of the sample, S2 = 0.36 kg; and the significance level, α =
0.01. The null and alternate hypotheses are:
Ho : σ2≥ 0.64
H1 : σ2‹ 0.64
The chi-square statistic to test the variance is:
(n−1) S 2
χ 2= [with (n-1) degrees of freedom]
σ2
( 8−1 ) X 0.36
¿
0.64

=3.938(with 7 degrees of freedom)

The table chi-square value with 7 degrees of freedom and at the given significance level
of 0.01 placed at the left tail (0.99 from right tail) is 1.239, which is shown in Fig.
Since the calculated chi-square value (3.938) is more than the table chi-square left value
(1.239) for the given significance level of 0.01, it falls in the acceptance region. Hence H 0
is to be accepted.
Hence, we find that the variance of weight of cement bags is more than the specified
variance. Hence, the quality assistant should instruct the production supervisor to adjust
the cement filling machine to avoid customer complaints in future.

Two-tailed Chi-square Test Concerning Single Population Variance

8.31 The weight of a drug produced by Ganga Pharmaceuticals Company follows normal
distribution. The specified variance of the weight of the drug of this population is
0.25mg. The quality engineer of the firm claims that the variance of the weight of the
drug does not differ significantly from the specified variance of the weight of the drug
population. So, the purchase manager of Alpha Hospital who places order for that drug
with the Ganga Pharmaceuticals Company has selected a random sample of 12 drugs.
The variance of the weight of the sample is found to be 0.49 mg. Verify the intuition of
the quality manager of Ganga Pharmaceutical Company at a significance level of 0.10.
Solution. Weight of the drug follows normal distribution. We also have population
variance of the weight of drugs σ 2= 0.25 mg; sample size, n=12; sample variance of the
weight of drugs, S2 = 0.49 mg; and significance level, α =0.10.
Null and alternate hypotheses are:
Ho : σ2= 0.25
H1 : σ2≠0.25

The chi-square statistic to test the variance is:

(n−1) S 2
χ 2=
σ2
(12−1 ) X 0.49
¿
0.25

= 21.56(with 7 degrees of freedom)

2 2
The table chi-square varies with 11 degrees of freedom, χ α2 placed at left tail and χ α2
placed at right tail are 4.575 and 19.675, respectively, which are as shown in Fig.

Since the calculated chi-square value (21.56) is more than the table chi-square value at
the right tail (19.675), it falls in the rejection region. Hence, H 0 is to be rejected.
Therefore, the variance of the weight of the drug is different from the specified
variance. Hence, the quality engineer should take steps to adjust the process to bring
the variance of the weight of the drug within the limit.

One-tailed F-tests Concerning Two Population Variances

8.32 Two independent samples of students representing Region-1 and Region-2 of a
programme under the distance education of a university are taken from normal
populations with the same variance. The variance of the marks of 10 students of sample
taken from the Region-1 is 64
And the variance of the marks of 15 students of another sample taken from the Region-
2 is 40. Check whether the variance of the marks of the first sample is less than the
variance of the marks of the second sample at a significance level of 1.05.
Solution. We have
Size of the sample from the Region-1, n1 =10
Size of the sample from the Region-2, n2 =15
Variance of the sample from the Region-1, S12 = 64
Variance of the sample from the Region-2, S22 = 40
Significance level, α = 0.05
Hypotheses to be tested are:
H0:σ 12 ≤ σ 22
H1:σ 12> σ 22
In this example, σ 12=σ 22. Hence the F-statistic is computed as:
S 21
F= 2
S2
The table F value with (9, 14) degrees of freedom at the given significance level of 0.05
is 2.65, which is shown in Fig. along with the region of acceptance and the region of
rejection.

Since the calculated F ratio (1.6) is less than the table F value (2.65), it falls in the
acceptance region. Hence, H0 is to be accepted. Therefore, the variance of the marks of
the students in the Region-1 is less than the variance of the marks of the students in the
Region-2.