Question:

A company has two plants to manufacture cars. Plant A manufactures 70% of the cars and
Plant B manufactures 30%. At Plant A, 80% of the cars produced are of standard quality, and at
Plant B, 90% of the cars produced are of standard quality. A car is picked up at random and is
found to be of standard quality. What is the probability that it has come from Plant B?

Solution Approach:
This is a conditional probability problem, and we can use Bayes' Theorem to solve it. The
formula for Bayes' Theorem is:

P(B∣S)=P(S∣B)⋅P(B)P(S)P(B|S) = \frac{P(S|B) \cdot P(B)}

{P(S)}P(B∣S)=P(S)P(S∣B)⋅P(B)

Where:

● P(B∣S)P(B|S)P(B∣S) is the probability that the car came from Plant B given
that it is of standard quality.
● P(S∣B)P(S|B)P(S∣B) is the probability that a car from Plant B is of standard
quality.
● P(B)P(B)P(B) is the probability that a car was produced by Plant B.
● P(S)P(S)P(S) is the total probability that a car is of standard quality (from either Plant A
or Plant B).

Step 1: Find the probabilities

From the question:

● P(B)=0.30P(B) = 0.30P(B)=0.30 (since 30% of cars are from Plant B),

● P(A)=0.70P(A) = 0.70P(A)=0.70 (since 70% of cars are from Plant A),
● P(S∣B)=0.90P(S|B) = 0.90P(S∣B)=0.90 (since 90% of cars from Plant B are
standard quality),
● P(S∣A)=0.80P(S|A) = 0.80P(S∣A)=0.80 (since 80% of cars from Plant A are
standard quality).

Step 2: Find the total probability P(S)P(S)P(S)

The total probability that a randomly selected car is of standard quality is:

P(S)=P(S∣A)⋅P(A)+P(S∣B)⋅P(B)P(S) = P(S|A) \cdot P(A) + P(S|B) \cdot

P(B)P(S)=P(S∣A)⋅P(A)+P(S∣B)⋅P(B) P(S)=(0.80×0.70)+(0.90×0.30)P(S) = (0.80 \
times 0.70) + (0.90 \times 0.30)P(S)=(0.80×0.70)+(0.90×0.30)
P(S)=0.56+0.27=0.83P(S) = 0.56 + 0.27 = 0.83P(S)=0.56+0.27=0.83

Step 3: Apply Bayes' Theorem

Now, using Bayes' Theorem:
P(B∣S)=P(S∣B)⋅P(B)P(S)P(B|S) = \frac{P(S|B) \cdot P(B)}
{P(S)}P(B∣S)=P(S)P(S∣B)⋅P(B) P(B∣S)=0.90×0.300.83=0.270.83≈0.3253P(B|S) = \
frac{0.90 \times 0.30}{0.83} = \frac{0.27}{0.83} \approx
0.3253P(B∣S)=0.830.90×0.30=0.830.27≈0.3253

Final Answer:
The probability that the car came from Plant B given that it is of standard quality is
approximately 0.3253, or 32.53%.

Question:
The life time of a certain kind of battery has a mean life of 400 hours and a standard deviation of
45 hours. Assuming the distribution of life time to be normal, find:

(i) The percentage of batteries with a life time of at least 470 hours.

(ii) The proportion of batteries with a life time between 385 and 415 hours.

(iii) The minimum life of the best 5% of batteries.

Solution:
We will use the properties of the normal distribution, where:

● Mean (μ\muμ) = 400 hours

● Standard deviation (σ\sigmaσ) = 45 hours

We can use the Z-score formula to standardize the data:

Z=X−μσZ = \frac{X - \mu}{\sigma}Z=σX−μ

Where XXX is the value we are interested in, μ\muμ is the mean, and σ\sigmaσ is the standard
deviation.

(i) The percentage of batteries with a life time of at least 470 hours.
First, we calculate the Z-score for X=470X = 470X=470 hours:
Z=470−40045=7045≈1.556Z = \frac{470 - 400}{45} = \frac{70}{45} \approx
1.556Z=45470−400=4570≈1.556

Next, we look up the Z-score value of 1.556 in a standard normal distribution table (or use a
calculator). The area to the left of this Z-score (cumulative probability) is approximately 0.9406,
meaning 94.06% of the batteries have a life time less than 470 hours.

The percentage of batteries with a life time at least 470 hours is:

1−0.9406=0.0594or5.94%1 - 0.9406 = 0.0594 \quad \text{or} \quad

5.94\%1−0.9406=0.0594or5.94%

Thus, 5.94% of batteries have a life time of at least 470 hours.

(ii) The proportion of batteries with a life time between 385 and 415 hours.
We calculate the Z-scores for both values:

● For X=385X = 385X=385: Z=385−40045=−1545=−0.333Z = \frac{385 -

400}{45} = \frac{-15}{45} = -0.333Z=45385−400=45−15=−0.333
● For X=415X = 415X=415: Z=415−40045=1545=0.333Z = \frac{415 - 400}
{45} = \frac{15}{45} = 0.333Z=45415−400=4515=0.333

Looking up these Z-scores in the standard normal distribution table:

● For Z=−0.333Z = -0.333Z=−0.333, the cumulative probability is

approximately 0.3707.
● For Z=0.333Z = 0.333Z=0.333, the cumulative probability is approximately 0.6293.

The proportion of batteries with a life time between 385 and 415 hours is:

0.6293−0.3707=0.25860.6293 - 0.3707 = 0.25860.6293−0.3707=0.2586

Thus, the proportion of batteries with a life time between 385 and 415 hours is 25.86%.

(iii) The minimum life of the best 5% of batteries.

For the best 5% of batteries, we need to find the Z-score that corresponds to the upper 5% in a
standard normal distribution. From the Z-score table, the Z-score for 95% (since 100% - 5% =
95%) is approximately 1.645.

Now we use the Z-score formula to find the corresponding life time:

Z=X−40045Z = \frac{X - 400}{45}Z=45X−400 1.645=X−400451.645 = \frac{X -

400}{45}1.645=45X−400
Solving for XXX:

X=1.645×45+400=73.025+400=473.025X = 1.645 \times 45 + 400 = 73.025 + 400 =

473.025X=1.645×45+400=73.025+400=473.025

Thus, the minimum life of the best 5% of batteries is approximately 473.03 hours.

Final Answers:
● (i) The percentage of batteries with a life time of at least 470 hours: 5.94%
● (ii) The proportion of batteries with a life time between 385 and 415 hours: 25.86%
● (iii) The minimum life of the best 5% of batteries: 473.03 hours.

Question:
(b) What are the properties of Karl Pearson’s coefficient of correlation? Calculate the correlation
coefficient from the following information:

● N=10N = 10N=10
● ΣX=140\Sigma X = 140ΣX=140
● ΣY=150\Sigma Y = 150ΣY=150
● Σ(X−10)2=180\Sigma(X - 10)^2 = 180Σ(X−10)2=180
● Σ(Y−15)2=215\Sigma(Y - 15)^2 = 215Σ(Y−15)2=215
● Σ(X−10)(Y−15)=60\Sigma(X - 10)(Y - 15) = 60Σ(X−10)(Y−15)=60

Answer:
The properties of Karl Pearson’s coefficient of correlation, denoted as rrr, include the following:

1. Range: It ranges between -1 and 1, where:

○ r=1r = 1r=1 indicates a perfect positive correlation.
○ r=−1r = -1r=−1 indicates a perfect negative correlation.
○ r=0r = 0r=0 indicates no linear correlation.
2. Symmetry: The correlation between two variables XXX and YYY is the same regardless
of which variable is considered first, i.e., r(X,Y)=r(Y,X)r(X, Y) = r(Y, X)r(X,Y)=r(Y,X).
3. Unit-Free: The coefficient is dimensionless, meaning it does not depend on the units of
measurement of the variables.
4. Linear Relationship: The coefficient only measures the strength and direction of the
linear relationship between two variables.
5. Sensitive to Outliers: Pearson’s correlation can be greatly affected by outliers in the
data.
Calculation of Karl Pearson’s Correlation Coefficient rrr:

The formula for the correlation coefficient rrr is:

r=Σ(X−Xˉ)(Y−Yˉ)Σ(X−Xˉ)2⋅Σ(Y−Yˉ)2r = \frac{\Sigma(X - \bar{X})(Y - \bar{Y})}{\

sqrt{\Sigma(X - \bar{X})^2 \cdot \Sigma(Y - \bar{Y})^2}}r=Σ(X−Xˉ)2⋅Σ(Y−Yˉ)2
Σ(X−Xˉ)(Y−Yˉ)

Given the data:

● N=10N = 10N=10
● Σ(X−10)2=180\Sigma(X - 10)^2 = 180Σ(X−10)2=180
● Σ(Y−15)2=215\Sigma(Y - 15)^2 = 215Σ(Y−15)2=215
● Σ(X−10)(Y−15)=60\Sigma(X - 10)(Y - 15) = 60Σ(X−10)(Y−15)=60

Substitute the values into the formula:

r=60180×215=6038700=60196.73≈0.305r = \frac{60}{\sqrt{180 \times 215}}

= \frac{60}{\sqrt{38700}} = \frac{60}{196.73} \approx 0.305r=180×21560
=3870060=196.7360≈0.305

Thus, the correlation coefficient rrr is approximately 0.305, indicating a moderate positive linear
relationship between the variables.

Question:
(b) Find the correlation between age and the playing habits of the following students:

Age (Years) No. of Regular Players

Students

15 250 200

16 200 150

17 150 90

18 120 48

19 100 30
 20 80 12

Solution:
To find the correlation between age and the number of regular players, we can use Karl
Pearson's coefficient of correlation formula:

r=n∑XY−(∑X)(∑Y)[n∑X2−(∑X)2][n∑Y2−(∑Y)2]r = \frac{n \sum XY - (\sum X)(\sum

Y)}{\sqrt{[n \sum X^2 - (\sum X)^2][n \sum Y^2 - (\sum Y)^2]}}r=[n∑X2−(∑X)2]
[n∑Y2−(∑Y)2]n∑XY−(∑X)(∑Y)

Where:

● XXX represents the ages of the students.

● YYY represents the number of regular players.
● nnn is the number of pairs of data.

We'll calculate the necessary values step by step:

Age (X) Regular Players (Y) X2X^ Y2Y^ XYXY

2X2 2Y2 XY

15 200 225 40000 3000

16 150 256 22500 2400

17 90 289 8100 1530

18 48 324 2304 864

19 30 361 900 570

20 12 400 144 240

Now, calculate the sums:

● ∑X=15+16+17+18+19+20=105\sum X = 15 + 16 + 17 + 18 + 19 + 20 =
105∑X=15+16+17+18+19+20=105
● ∑Y=200+150+90+48+30+12=530\sum Y = 200 + 150 + 90 + 48 + 30 + 12
= 530∑Y=200+150+90+48+30+12=530
● ∑X2=225+256+289+324+361+400=1855\sum X^2 = 225 + 256 + 289 +
324 + 361 + 400 = 1855∑X2=225+256+289+324+361+400=1855
● ∑Y2=40000+22500+8100+2304+900+144=73848\sum Y^2 = 40000 +
22500 + 8100 + 2304 + 900 + 144 =
73848∑Y2=40000+22500+8100+2304+900+144=73848
 ● ∑XY=3000+2400+1530+864+570+240=8604\sum XY = 3000 + 2400 +
1530 + 864 + 570 + 240 =
8604∑XY=3000+2400+1530+864+570+240=8604
● n=6n = 6n=6

Now substitute these values into the formula for rrr:

r=6(8604)−(105)(530)[6(1855)−(105)2][6(73848)−(530)2]r = \frac{6(8604) - (105)

(530)}{\sqrt{[6(1855) - (105)^2][6(73848) - (530)^2]}}r=[6(1855)−(105)2]
[6(73848)−(530)2]6(8604)−(105)(530)

Calculate step by step:

1. 6(8604)=516246(8604) = 516246(8604)=51624
2. (105)(530)=55650(105)(530) = 55650(105)(530)=55650
3. 6(1855)=111306(1855) = 111306(1855)=11130
4. (105)2=11025(105)^2 = 11025(105)2=11025
5. 6(73848)=4430886(73848) = 4430886(73848)=443088
6. (530)2=280900(530)^2 = 280900(530)2=280900

Now:

r=51624−55650[11130−11025][443088−280900]r = \frac{51624 - 55650}{\

sqrt{[11130 - 11025][443088 - 280900]}}r=[11130−11025][443088−280900]
51624−55650

Simplifying:

r=−4026(105)(162188)r = \frac{-4026}{\sqrt{(105)(162188)}}r=(105)(162188)
−4026 r=−402617029740r = \frac{-4026}{\sqrt{17029740}}r=17029740−4026
r=−40264129.12r = \frac{-4026}{4129.12}r=4129.12−4026 r≈−0.975r \approx -
0.975r≈−0.975

Conclusion:
The correlation coefficient rrr is approximately -0.975, indicating a strong negative correlation
between age and the number of regular players. As the age of students increases, the number
of regular players tends to decrease significantly.

Question:
The coefficient of correlation between XXX and YYY for 20 items is 0.3. The mean of XXX is 15
and that of YYY is 20, while standard deviations are 4 and 5 respectively. At the time of
calculation, one item 27 has wrongly been taken as 17 in case of XXX series and one item 35
was wrongly taken instead of 30 in case of YYY series. Find the correct coefficient of
correlation.

Solution Approach:

Step 1: Sum of XXX and YYY using their means

We know that:

Sum of Xold=Xˉ×n=15×20=300\text{Sum of } X_{\text{old}} = \bar{X} \times n = 15 \times 20 =

300Sum of Xold=Xˉ×n=15×20=300 Sum of Yold=Yˉ×n=20×20=400\text{Sum of } Y_{\text{old}} =
\bar{Y} \times n = 20 \times 20 = 400Sum of Yold=Yˉ×n=20×20=400

Step 2: Correct the sums for XXX and YYY

Since one item was wrongly taken as 17 instead of 27 in XXX, and another item was taken as
35 instead of 30 in YYY, we adjust the sums accordingly:

Sum of Xnew=Sum of Xold+(27−17)=300+10=310\text{Sum of } X_{\text{new}}

= \text{Sum of } X_{\text{old}} + (27 - 17) = 300 + 10 = 310Sum of Xnew=Sum
of Xold+(27−17)=300+10=310 Sum of Ynew=Sum of Yold+(30−35)=400−5=395\
text{Sum of } Y_{\text{new}} = \text{Sum of } Y_{\text{old}} + (30 - 35) = 400 -
5 = 395Sum of Ynew=Sum of Yold+(30−35)=400−5=395

Step 3: Correct the sum of products

Now, let's adjust the sum of products. The incorrect contribution to the sum of products is
calculated as:

Incorrect product=(Xincorrect−Xˉ)×(Yincorrect−Yˉ)=(17−15)×(35−20)=2×15=30\
text{Incorrect product} = (X_{\text{incorrect}} - \bar{X}) \times (Y_{\
text{incorrect}} - \bar{Y}) = (17 - 15) \times (35 - 20) = 2 \times 15 = 30Incorrect
product=(Xincorrect−Xˉ)×(Yincorrect−Yˉ)=(17−15)×(35−20)=2×15=30

The correct contribution to the sum of products is:

Correct product=(Xcorrect−Xˉ)×(Ycorrect−Yˉ)=(27−15)×(30−20)=12×10=120\
text{Correct product} = (X_{\text{correct}} - \bar{X}) \times (Y_{\text{correct}} -
\bar{Y}) = (27 - 15) \times (30 - 20) = 12 \times 10 = 120Correct
product=(Xcorrect−Xˉ)×(Ycorrect−Yˉ)=(27−15)×(30−20)=12×10=120

The change in the sum of products is:

Change in sum of products=120−30=90\text{Change in sum of products} = 120 -

30 = 90Change in sum of products=120−30=90
Step 4: Calculate the old sum of products
The sum of products (Σ(Xi−Xˉ)(Yi−Yˉ)\Sigma (X_i - \bar{X})(Y_i - \bar{Y})Σ(Xi−Xˉ)
(Yi−Yˉ)) can be derived from the old correlation formula:

rold=Σ(Xi−Xˉ)(Yi−Yˉ)nσXσYr_{\text{old}} = \frac{\Sigma (X_i - \bar{X})(Y_i - \

bar{Y})}{n \sigma_X \sigma_Y}rold=nσXσYΣ(Xi−Xˉ)(Yi−Yˉ)

Rearranging this formula, we get:

Σ(Xi−Xˉ)(Yi−Yˉ)=rold×(nσXσY)=0.3×(20×4×5)=0.3×400=120\Sigma (X_i - \
bar{X})(Y_i - \bar{Y}) = r_{\text{old}} \times (n \sigma_X \sigma_Y) = 0.3 \times
(20 \times 4 \times 5) = 0.3 \times 400 = 120Σ(Xi−Xˉ)(Yi−Yˉ)=rold×(nσXσY
)=0.3×(20×4×5)=0.3×400=120

Step 5: Calculate the new sum of products

The new sum of products is:

Σnew(Xi−Xˉ)(Yi−Yˉ)=Σold(Xi−Xˉ)(Yi−Yˉ)+Change in sum of products\Sigma_{\

text{new}} (X_i - \bar{X})(Y_i - \bar{Y}) = \Sigma_{\text{old}} (X_i - \bar{X})(Y_i -
\bar{Y}) + \text{Change in sum of products}Σnew(Xi−Xˉ)(Yi−Yˉ)=Σold(Xi−Xˉ)(Yi
−Yˉ)+Change in sum of products =120+90=210= 120 + 90 = 210=120+90=210

Step 6: Calculate the new correlation coefficient

Now, using the formula for the correlation coefficient again:

rnew=Σnew(Xi−Xˉ)(Yi−Yˉ)nσXσYr_{\text{new}} = \frac{\Sigma_{\text{new}} (X_i

- \bar{X})(Y_i - \bar{Y})}{n \sigma_X \sigma_Y}rnew=nσXσYΣnew(Xi−Xˉ)(Yi−Yˉ)
rnew=21020×4×5=210400=0.525r_{\text{new}} = \frac{210}{20 \times 4 \times
5} = \frac{210}{400} = 0.525rnew=20×4×5210=400210=0.525

Final Answer:
The correct coefficient of correlation is 0.525.

Question:
(c) Given below is the monthly trend equation:

Yc=130+1.8XY_c = 130 + 1.8XYc=130+1.8X

[Origin: 1/7/2002, XXX unit = one month, and YYY unit = monthly sales]
Convert the above equation to an annual trend equation and estimate the sales for the year
2010.

Answer:
Step 1: Convert the monthly trend equation to an annual trend equation

The given monthly trend equation is:

Yc=130+1.8XY_c = 130 + 1.8XYc=130+1.8X

Where:

● YcY_cYc = monthly sales

● XXX = the number of months from the origin (1/7/2002)

To convert this into an annual trend equation, we recognize that there are 12 months in a year.
Let XannualX_{\text{annual}}Xannual represent the number of years from the origin. Since each
year has 12 months, we can write:

Xmonthly=12×XannualX_{\text{monthly}} = 12 \times X_{\text{annual}}Xmonthly=12×Xannual

Substitute this into the monthly trend equation:

Yannual=130+1.8(12×Xannual)Y_{\text{annual}} = 130 + 1.8(12 \times X_{\

text{annual}})Yannual=130+1.8(12×Xannual)

Simplify:

Yannual=130+21.6XannualY_{\text{annual}} = 130 + 21.6X_{\text{annual}}Yannual

=130+21.6Xannual

Thus, the annual trend equation becomes:

Yannual=130+21.6XannualY_{\text{annual}} = 130 + 21.6X_{\text{annual}}Yannual

=130+21.6Xannual

Step 2: Estimate the sales for the year 2010

To find the sales for the year 2010, we first calculate how many years have passed since the
origin (1/7/2002).

From 1/7/2002 to 1/1/2010 is:

2010−2002=8 years2010 - 2002 = 8 \text{ years}2010−2002=8 years

Now, substitute Xannual=8X_{\text{annual}} = 8Xannual=8 into the annual trend equation:

Y2010=130+21.6×8Y_{\text{2010}} = 130 + 21.6 \times 8Y2010=130+21.6×8

Y2010=130+172.8=302.8Y_{\text{2010}} = 130 + 172.8 = 302.8Y2010=130+172.8=302.8
Therefore, the estimated sales for the year 2010 are 302.8 units.

Question:
Below are the figures of production (in '000 tons) of a factory:

Year Production ('000

tons)

2004 77

2005 78

2006 94

2007 85

2008 91

2009 98

2010 90

(i) Fit a straight-line trend by the method of least squares.

(ii) What is the monthly increase in production?

(i) Fit a straight-line trend by the method of least squares.

The equation of a straight-line trend is: Y=a+bXY = a + bXY=a+bX

Where:

● YYY is the production.

● XXX is the time (years).
● aaa and bbb are the coefficients to be determined using least squares.

Steps to compute aaa and bbb:

1. Create the year variable as XXX, but simplify by using deviations from the
midpoint year (2007). So we will use X=−3,−2,−1,0,1,2,3X = -3, -2, -1, 0, 1,
2, 3X=−3,−2,−1,0,1,2,3.
 2. Compute ΣX\Sigma XΣX, ΣY\Sigma YΣY, ΣXY\Sigma XYΣXY, and ΣX2\Sigma X^2ΣX2
for the given values.
3. Use the formulas:
○ b=ΣXYΣX2b = \frac{\Sigma XY}{\Sigma X^2}b=ΣX2ΣXY
○ a=ΣYn−bΣXna = \frac{\Sigma Y}{n} - b \frac{\Sigma X}{n}a=nΣY
−bnΣX
4. Where n is the number of years (7 in this case).

Let me compute that.

The equation of the straight-line trend is:

Y=87.57+2.71XY = 87.57 + 2.71XY=87.57+2.71X

Where XXX represents the time in years (deviations from 2007).

(ii) What is the monthly increase in production?

The slope of the trend line (b=2.71b = 2.71b=2.71) represents the yearly increase in production.
To find the monthly increase:

Monthly increase=b12=2.7112=0.226 thousand tons per month.\text{Monthly increase} = \

frac{b}{12} = \frac{2.71}{12} = 0.226 \text{ thousand tons per month}.Monthly increase=12b
=122.71=0.226 thousand tons per month.

So, the monthly increase in production is approximately 0.226 thousand tons (or 226 tons).

Question:
Given the following data:

Measures Advertising Expenditure (₹ lakh) Sales (₹

crore)

Mean 40 90

Standard Deviation (σ) 10 20

The coefficient of correlation is +0.92.

(i) Calculate the two regression equations.

(ii) Estimate the likely sales for an advertising expenditure of ₹ 60 lakh.

(iii) What should be the advertising expenditure for attaining sales of ₹ 120 crore?

Solution:
(i) Regression Equations:

1. Regression Equation of Sales on Advertising (Y on X):

Y−Y‾=bYX(X−X‾)Y - \overline{Y} = b_{YX}(X - \overline{X})Y−Y=bYX(X−X)
Where:
○ X‾=40\overline{X} = 40X=40 (mean advertising expenditure in ₹ lakh)
○ Y‾=90\overline{Y} = 90Y=90 (mean sales in ₹ crore)
○ σX=10\sigma_X = 10σX=10 (standard deviation of advertising expenditure)
○ σY=20\sigma_Y = 20σY=20 (standard deviation of sales)
○ r=0.92r = 0.92r=0.92 (correlation coefficient)
2. The regression coefficient bYXb_{YX}bYX is calculated as:
bYX=r×σYσX=0.92×2010=1.84b_{YX} = r \times \frac{\sigma_Y}{\sigma_X}
= 0.92 \times \frac{20}{10} = 1.84bYX=r×σXσY=0.92×1020=1.84
So, the regression equation of sales on advertising is:
Y−90=1.84(X−40)Y - 90 = 1.84(X - 40)Y−90=1.84(X−40)
Simplifying:
Y=1.84X−73.6+90Y = 1.84X - 73.6 + 90Y=1.84X−73.6+90 Y=1.84X+16.4Y
= 1.84X + 16.4Y=1.84X+16.4
3. Regression Equation of Advertising on Sales (X on Y):
X−X‾=bXY(Y−Y‾)X - \overline{X} = b_{XY}(Y - \overline{Y})X−X=bXY(Y−Y)
The regression coefficient bXYb_{XY}bXY is calculated as:
bXY=r×σXσY=0.92×1020=0.46b_{XY} = r \times \frac{\sigma_X}{\sigma_Y}
= 0.92 \times \frac{10}{20} = 0.46bXY=r×σYσX=0.92×2010=0.46
So, the regression equation of advertising on sales is:
X−40=0.46(Y−90)X - 40 = 0.46(Y - 90)X−40=0.46(Y−90)
Simplifying:
X=0.46Y−41.4+40X = 0.46Y - 41.4 + 40X=0.46Y−41.4+40 X=0.46Y−1.4X =
0.46Y - 1.4X=0.46Y−1.4

(ii) Estimate the likely sales for an advertising expenditure of ₹ 60 lakh:

Using the regression equation of sales on advertising:

Y=1.84X+16.4Y = 1.84X + 16.4Y=1.84X+16.4

Substitute X=60X = 60X=60:

Y=1.84(60)+16.4=110.4+16.4=126.8Y = 1.84(60) + 16.4 = 110.4 + 16.4 =
126.8Y=1.84(60)+16.4=110.4+16.4=126.8

Thus, the likely sales for an advertising expenditure of ₹ 60 lakh is ₹ 126.8 crore.

(iii) What should be the advertising expenditure for attaining sales of ₹ 120 crore?

Using the regression equation of advertising on sales:

X=0.46Y−1.4X = 0.46Y - 1.4X=0.46Y−1.4

Substitute Y=120Y = 120Y=120:

X=0.46(120)−1.4=55.2−1.4=53.8X = 0.46(120) - 1.4 = 55.2 - 1.4 =

53.8X=0.46(120)−1.4=55.2−1.4=53.8

Thus, the required advertising expenditure to attain sales of ₹ 120 crore is ₹ 53.8 lakh.