Decision Science

Q1. Draw the decision tree diagram and explain the best possible decision based on EMVs (expected
Monetary Values) On the 1st April 2023, Rajinder Saproo, an investor is in a dilemma for the
investment of 10 lakh INR. He has consulted his Mumbai based friend, MukulBhai Gadhecha, an
investment expert for this matter. Mr Saproo went to meet him at his office located at Prabhadevi.
Mukulbhai asked him to express his feeling for the market situation in coming economic year.

Mr Saproo assumed the economic growth for the coming year in the following way. 10 % optimist for
the ‘Good Economic Growth’ 50 % optimist for the ‘Moderate Economic Growth’ 40 % optimist for
the ‘Lower Economic Growth’. In addition to this, MuKulbhai Gadhecha carried out his analysis to
derive possible payoff values considering the various investment options as per the economic
situations (presented in the table given below). Figures are in INR on the investment of whole
amount of 10 lakhs.

If you were in place of Mukulbhai Gadhecha, What do you recommend to Mr. Saproo? Note: You are
not allowed to use any software for the calculation. But for the preparation of Decision tree diagram
you are advised to use Software like MS -Word, PowerPoint etc. (10 Marks)

Answer- Title: Investment Recommendation for Mr. Saproo Using Expected Monetary Values (EMVs)

Introduction:

Mr. Rajinder Saproo, an investor with 10 lakh INR, sought the know-how of MukulbhaiGadhecha, an
investment expert based in Mumbai. Mukulbhai became tasked with advising Mr. Saproo on the
optimal investment Decision given sure economic growth scenarios. Mr. Saproo's notion of the
financial landscape for the upcoming year became divided into three procedures: 10% optimistic
about 'true economic growth,' 50% confident about 'moderate economic increase,' and 40% happy
about 'lower financial increase.' Mukulbhai Gadhecha comprehensively analyzed potential payoffs for
diverse investment alternatives underneath these situations. The investment alternatives under
attention had been Maruti Suzuki shares, Tata Motor stocks, and D Mart stocks, each supplying
exclusive returns relying on the financial situation.
Decision Tree Diagram: To provide Mr. Saproo with clean and informed advice, we built a decision
tree diagram that visually represents the investment alternatives and the associated chances of every
economic scenario. The decision tree enables demonstrating the ability outcomes and anticipated
returns below one-of-a-kind instances. The following is a Decision Tree Diagram.

Investment Decision (10, 00,000 INR)

/ | \

/ | \

Good Economic Growth Moderate Economic Growth \ Lower Economic Growth

/ \

10% (0.10) / 50% (0.50) 40% (0.40)

/ \ /

/ \ /

/ \ /

/ \ /

Mauti Suzuki Shares TATA Motor Shares D Mart shares

3, 00,000 (0.10) 4, 00,000 (0.10) 4,50,000 (0.10)

| | |

1, 20,000 (0.50) 1,00,000 (0.50) 2,30,000 (0.50)

| | |

50,000 (0.40) 10,000 (0.40) 30,000 (0.40)

The Decision tree depicts Mr. Saproo's picks based totally on his to-be-had funding of 10, 00,000 INR
and the probabilities assigned to each economic growth scenario. The three investment options are
laid out on the final degree of the tree, each with its respective returns for 'true financial increase,'
'moderate financial growth,' and 'decreased financial increase.'

Expected Monetary Values (EMVs):

To decide the satisfactory funding decision for Mr. Saproo, we calculated the expected economic Values
(EMVs) for each funding option under every economic scenario. EMV is a crucial Decision-making
device because it considers both capacity returns and their associated possibilities. EMV is calculated
as the sum of every possible final result's product and probability.

For Mauti Suzuki (MS) Shares:

• Good Economic Growth (10%): EMV = 10% * 3,00,000 = 30,000

• Moderate Economic Growth (50%): EMV = 50% * 50,000 = 25,000

• Lower Economic Growth (40%): EMV = 40% * 50,000 = 20,000

Total EMV for MS Shares: 30,000 + 25,000 + 20,000 = 75,000 INR

For TATA Motor Shares:

• Good Economic Growth (10%): EMV = 10% * 4,00,000 = 40,000

• Moderate Economic Growth (50%): EMV = 50% * 1,00,000 = 50,000

• Lower Economic Growth (40%): EMV = 40% * 10,000 = 4,000

Total EMV for TATA Motor Shares: 40,000 + 50,000 + 4,000 = 94,000 INR

For D Mart Shares:

• Good Economic Growth (10%): EMV = 10% * 4,50,000 = 45,000

• Moderate Economic Growth (50%): EMV = 50% * 2,30,000 = 1,15,000

• Lower Economic Growth (40%): EMV = 40% * 30,000 = 12,000

Total EMV for D Mart Shares: 45,000 + 1,15,000 + 12,000 = 1,72,000 INR

Based on the Expected Monetary Values (EMVs) calculated for each investment option, Mr. Saproo
should consider investing in D Mart Shares as they have the highest EMV of 1,72,000 INR. This
recommendation is based on the assumption of different economic scenarios and their respective
probabilities, making it the most financially favorable choice among the given options.

Recommendation:

Primarily based on the calculated EMVs for every investment Decision, the excellent investment
preference for Mr. Saproo is D Mart shares. This feature gives the highest predicted go-back for many
of the three investment picks, with an EMV of approximately 172,000 INR.

D Mart shares align with Mr. Saproo's goal of maximizing ability returns even as thinking about the
related chance. This recommendation is made after a radical evaluation of the predicted returns below
various financial growth scenarios, as pondered inside the EMVs.

Conclusion:

In conclusion, while offering to invest his 10 lakh INR, Mr. Saproo must opt for D Mart shares. This
desire is supported by carefully considering probabilities and potential returns for every funding
option. Through specializing in D Mart shares, Mr. Saproo can function himself to maximize his returns
in keeping with his optimism about the monetary future, as indicated by the probabilities he assigned
to different economic growth scenarios.

However, Mr. Saproo must keep in mind that all investments bring some level of risk, and he needs to
continue to reveal market conditions and periodically review his investment portfolio to make
informed selections as situations change. This recommendation is primarily based on the available
information and the understanding of economic scenarios at the time of the decision.

Q2. Draw the appropriate chart using MS EXCEL, also carryout Exponential Smoothing model in MS
EXCEL for the given Alpha values, copy the results (from EXCEL to Word) and explain which alpha is
relatively better for the forecast. In this journey of identification of relatively better way of forecasting
you may take the support of MAD and MSE type of errors. Suggested Alpha values are 0.1, 0.2, 0.5,
0.7, 0.9. (10 Marks)

Answer- The Quest for the Optimal Alpha:

Inside the realm of forecasting, we embark on a quest to unveil the alpha value that holds the key to
superior predictive powers. Our journey takes us through the dense forest of mean Absolute Deviation
(MAD) and the treacherous terrain of suggest-squared mistakes (MSE). These metrics will be our
guiding stars in this odyssey.

The Contestants:

Our noble contestants are alpha values—0.1, 0.2, 0.5, 0.7, and 0.9. each Alpha seeks to prove its worth
in the forecasting world, but only one shall emerge victorious.

Alpha = 0.1

(3)
(1) (2)
Exponential Smoothing
year Sales
(α=0.1)

1 977 977

2 1127 0.1⋅977+0.9⋅977=977

3 694 0.1⋅1127+0.9⋅977=992

4 1357 0.1⋅694+0.9⋅992=962.2

5 1020 0.1⋅1357+0.9⋅962.2=1001.68

6 1187 0.1⋅1020+0.9⋅1001.68=1003.512

7 866 0.1⋅1187+0.9⋅1003.512=1021.8608

8 1459 0.1⋅866+0.9⋅1021.8608=1006.2747

9 1163 0.1⋅1459+0.9⋅1006.2747=1051.5472

10 991 0.1⋅1163+0.9⋅1051.5472=1062.6925

11 1411 0.1⋅991+0.9⋅1062.6925=1055.5233

12 1323 0.1⋅1411+0.9⋅1055.5233=1091.0709

13 995 0.1⋅1323+0.9⋅1091.0709=1114.2638

14 1764 0.1⋅995+0.9⋅1114.2638=1102.3375

15 1552 0.1⋅1764+0.9⋅1102.3375=1168.5037

16 1465 0.1⋅1552+0.9⋅1168.5037=1206.8533

17 1398 0.1⋅1465+0.9⋅1206.8533=1232.668
18 1893 0.1⋅1398+0.9⋅1232.668=1249.2012

19 1422 0.1⋅1893+0.9⋅1249.2012=1313.5811

20 2063 0.1⋅1422+0.9⋅1313.5811=1324.423

21 1703 0.1⋅2063+0.9⋅1324.423=1398.2807

22 1758 0.1⋅1703+0.9⋅1398.2807=1428.7526

23 0.1⋅1758+0.9⋅1428.7526=1461.6774

(1) (2) (3) (7)

(4) (5) (6)
yea Sale Exponential |%Error
Error |Error| Error2
r s Smoothing |

1 977 977

2 1127 977 1127-977=150 150 22500 13.31%

3 694 992 694-992=-298 298 88804 42.94%

4 1357 962.2 1357-962.2=394.8 394.8 155867.04 29.09%

5 1020 1001.68 1020-1001.68=18.32 18.32 335.6224 1.8%

6 1187 1003.512 1187-1003.512=183.488 183.488 33667.8461 15.46%

866-1021.8608=-
7 866 1021.8608 155.8608 24292.589 18%
155.8608

1459-
8 1459 1006.2747 452.7253 204960.1792 31.03%
1006.2747=452.7253

1163-
9 1163 1051.5472 111.4528 12421.7159 9.58%
1051.5472=111.4528

10 991 1062.6925 991-1062.6925=-71.6925 71.6925 5139.8179 7.23%

1411-
11 1411 1055.5233 355.4767 126363.7049 25.19%
1055.5233=355.4767

1323-
12 1323 1091.0709 231.9291 53791.0871 17.53%
1091.0709=231.9291
 995-1114.2638=-
13 995 1114.2638 119.2638 14223.8658 11.99%
119.2638

1764-
14 1764 1102.3375 661.6625 437797.3109 37.51%
1102.3375=661.6625

1552-
15 1552 1168.5037 383.4963 147069.3983 24.71%
1168.5037=383.4963

1465-
16 1465 1206.8533 258.1467 66639.6949 17.62%
1206.8533=258.1467

17 1398 1232.668 1398-1232.668=165.332 165.332 27334.6664 11.83%

1893-
18 1893 1249.2012 643.7988 414476.8814 34.01%
1249.2012=643.7988

1422-
19 1422 1313.5811 108.4189 11754.6602 7.62%
1313.5811=108.4189

20 2063 1324.423 2063-1324.423=738.577 738.577 545496.0138 35.8%

1703-
21 1703 1398.2807 304.7193 92853.8625 17.89%
1398.2807=304.7193

1758-
22 1758 1428.7526 329.2474 108403.8411 18.73%
1428.7526=329.2474

6136.407 2594193.797
23 1461.6774 Total 428.88%
9 8

Forecasting errors
1. Mean absolute error (MAE), also called mean absolute deviation (MAD)
𝟔𝟏𝟑𝟔. 𝟒𝟎𝟕𝟗
MAD = = 𝟐𝟗𝟐. 𝟐𝟎𝟗𝟗
𝟐𝟏

2. Mean squared error (MSE)

2594193.7978
𝑀𝑆𝐸 = = 123533.038
21
 (3)
(1) (2)
Exponential Smoothing
year Sales
(α=0.2)

1 977 977

2 1127 0.2⋅977+0.8⋅977=977

3 694 0.2⋅1127+0.8⋅977=1007

4 1357 0.2⋅694+0.8⋅1007=944.4

5 1020 0.2⋅1357+0.8⋅944.4=1026.92

6 1187 0.2⋅1020+0.8⋅1026.92=1025.536

7 866 0.2⋅1187+0.8⋅1025.536=1057.8288

8 1459 0.2⋅866+0.8⋅1057.8288=1019.463

9 1163 0.2⋅1459+0.8⋅1019.463=1107.3704

10 991 0.2⋅1163+0.8⋅1107.3704=1118.4963

11 1411 0.2⋅991+0.8⋅1118.4963=1092.9971

12 1323 0.2⋅1411+0.8⋅1092.9971=1156.5977

13 995 0.2⋅1323+0.8⋅1156.5977=1189.8781

14 1764 0.2⋅995+0.8⋅1189.8781=1150.9025

15 1552 0.2⋅1764+0.8⋅1150.9025=1273.522

16 1465 0.2⋅1552+0.8⋅1273.522=1329.2176

17 1398 0.2⋅1465+0.8⋅1329.2176=1356.3741

18 1893 0.2⋅1398+0.8⋅1356.3741=1364.6993

19 1422 0.2⋅1893+0.8⋅1364.6993=1470.3594

20 2063 0.2⋅1422+0.8⋅1470.3594=1460.6875

21 1703 0.2⋅2063+0.8⋅1460.6875=1581.15

22 1758 0.2⋅1703+0.8⋅1581.15=1605.52
23 0.2⋅1758+0.8⋅1605.52=1636.016

(1) (2) (3) (7)

(4) (5) (6)
yea Sale Exponential |%Error
Error |Error| Error2
r s Smoothing |

1 977 977

2 1127 977 1127-977=150 150 22500 13.31%

3 694 1007 694-1007=-313 313 97969 45.1%

4 1357 944.4 1357-944.4=412.6 412.6 170238.76 30.41%

5 1020 1026.92 1020-1026.92=-6.92 6.92 47.8864 0.68%

6 1187 1025.536 1187-1025.536=161.464 161.464 26070.6233 13.6%

866-1057.8288=-
7 866 1057.8288 191.8288 36798.2885 22.15%
191.8288

8 1459 1019.463 1459-1019.463=439.537 439.537 193192.7392 30.13%

9 1163 1107.3704 1163-1107.3704=55.6296 55.6296 3094.6488 4.78%

991-1118.4963=-
10 991 1118.4963 127.4963 16255.3181 12.87%
127.4963

1411-
11 1411 1092.9971 318.0029 101125.8594 22.54%
1092.9971=318.0029

1323-
12 1323 1156.5977 166.4023 27689.7384 12.58%
1156.5977=166.4023

995-1189.8781=-
13 995 1189.8781 194.8781 37977.4851 19.59%
194.8781

1764-
14 1764 1150.9025 613.0975 375888.5406 34.76%
1150.9025=613.0975

15 1552 1273.522 1552-1273.522=278.478 278.478 77549.9951 17.94%

 1465-
16 1465 1329.2176 135.7824 18436.8596 9.27%
1329.2176=135.7824

17 1398 1356.3741 1398-1356.3741=41.6259 41.6259 1732.7171 2.98%

1893-
18 1893 1364.6993 528.3007 279101.6663 27.91%
1364.6993=528.3007

1422-1470.3594=-
19 1422 1470.3594 48.3594 2338.6328 3.4%
48.3594

2063-
20 2063 1460.6875 602.3125 362780.3118 29.2%
1460.6875=602.3125

21 1703 1581.15 1703-1581.15=121.85 121.85 14847.4167 7.16%

22 1758 1605.52 1758-1605.52=152.48 152.48 23250.1446 8.67%

5060.045 1888886.631
23 1636.016 Total 369%
4 7

Forecasting errors

1. Mean absolute error (MAE), also called mean absolute deviation (MAD)

5060.0454
𝑀𝐴𝐷 = = 240.9545
21

2. Mean squared error (MSE)

𝟏𝟖𝟖𝟖𝟖𝟖. 𝟔𝟑𝟏𝟕
MSE =. = 𝟖𝟗𝟗𝟒𝟔. 𝟗𝟖𝟐𝟓
𝟐𝟏

Alpha = 0.5

(3)
(1) (2)
Exponential Smoothing
year Sales
(α=0.5)

1 977 977

2 1127 0.5⋅977+0.5⋅977=977
3 694 0.5⋅1127+0.5⋅977=1052

4 1357 0.5⋅694+0.5⋅1052=873

5 1020 0.5⋅1357+0.5⋅873=1115

6 1187 0.5⋅1020+0.5⋅1115=1067.5

7 866 0.5⋅1187+0.5⋅1067.5=1127.25

8 1459 0.5⋅866+0.5⋅1127.25=996.625

9 1163 0.5⋅1459+0.5⋅996.625=1227.8125

10 991 0.5⋅1163+0.5⋅1227.8125=1195.4062

11 1411 0.5⋅991+0.5⋅1195.4062=1093.2031

12 1323 0.5⋅1411+0.5⋅1093.2031=1252.1016

13 995 0.5⋅1323+0.5⋅1252.1016=1287.5508

14 1764 0.5⋅995+0.5⋅1287.5508=1141.2754

15 1552 0.5⋅1764+0.5⋅1141.2754=1452.6377

16 1465 0.5⋅1552+0.5⋅1452.6377=1502.3188

17 1398 0.5⋅1465+0.5⋅1502.3188=1483.6594

18 1893 0.5⋅1398+0.5⋅1483.6594=1440.8297

19 1422 0.5⋅1893+0.5⋅1440.8297=1666.9149

20 2063 0.5⋅1422+0.5⋅1666.9149=1544.4574

21 1703 0.5⋅2063+0.5⋅1544.4574=1803.7287

22 1758 0.5⋅1703+0.5⋅1803.7287=1753.3644

23 0.5⋅1758+0.5⋅1753.3644=1755.6822

(1) (2) (3) (7)

(4) (5) (6)
yea Sale Exponential |%Error
Error |Error| Error2
r s Smoothing |
1 977 977

2 1127 977 1127-977=150 150 22500 13.31%

3 694 1052 694-1052=-358 358 128164 51.59%

4 1357 873 1357-873=484 484 234256 35.67%

5 1020 1115 1020-1115=-95 95 9025 9.31%

6 1187 1067.5 1187-1067.5=119.5 119.5 14280.25 10.07%

7 866 1127.25 866-1127.25=-261.25 261.25 68251.5625 30.17%

8 1459 996.625 1459-996.625=462.375 462.375 213790.6406 31.69%

9 1163 1227.8125 1163-1227.8125=-64.8125 64.8125 4200.6602 5.57%

10 991 1195.4062 991-1195.4062=-204.4062 204.4062 41781.915 20.63%

1411-
11 1411 1093.2031 317.7969 100994.8538 22.52%
1093.2031=317.7969

12 1323 1252.1016 1323-1252.1016=70.8984 70.8984 5026.5884 5.36%

13 995 1287.5508 995-1287.5508=-292.5508 292.5508 85585.9596 29.4%

1764-
14 1764 1141.2754 622.7246 387785.9391 35.3%
1141.2754=622.7246

15 1552 1452.6377 1552-1452.6377=99.3623 99.3623 9872.8676 6.4%

16 1465 1502.3188 1465-1502.3188=-37.3188 37.3188 1392.6964 2.55%

17 1398 1483.6594 1398-1483.6594=-85.6594 85.6594 7337.5369 6.13%

1893-
18 1893 1440.8297 452.1703 204457.9694 23.89%
1440.8297=452.1703

1422-1666.9149=-
19 1422 1666.9149 244.9149 59983.2867 17.22%
244.9149

2063-
20 2063 1544.4574 518.5426 268886.399 25.14%
1544.4574=518.5426

1703-1803.7287=-
21 1703 1803.7287 100.7287 10146.2738 5.91%
100.7287
22 1758 1753.3644 1758-1753.3644=4.6356 4.6356 21.4892 0.26%

5046.647 1877741.888
23 1755.6822 Total 388.09%
1 2

Forecasting errors

1. Mean absolute error (MAE), also called mean absolute deviation (MAD)
MAD =5046.6471 / 21=240.3165

2. Mean squared error (MSE)

MSE=1877741.8882 / 21=89416.2804

ALPHA = 0.7

(3)
(1) (2)
Exponential Smoothing
year Sales
(α=0.7)

1 977 977

2 1127 0.7⋅977+0.3⋅977=977

3 694 0.7⋅1127+0.3⋅977=1082

4 1357 0.7⋅694+0.3⋅1082=810.4

5 1020 0.7⋅1357+0.3⋅810.4=1193.02

6 1187 0.7⋅1020+0.3⋅1193.02=1071.906

7 866 0.7⋅1187+0.3⋅1071.906=1152.4718

8 1459 0.7⋅866+0.3⋅1152.4718=951.9415

9 1163 0.7⋅1459+0.3⋅951.9415=1306.8825

10 991 0.7⋅1163+0.3⋅1306.8825=1206.1647

11 1411 0.7⋅991+0.3⋅1206.1647=1055.5494
12 1323 0.7⋅1411+0.3⋅1055.5494=1304.3648

13 995 0.7⋅1323+0.3⋅1304.3648=1317.4094

14 1764 0.7⋅995+0.3⋅1317.4094=1091.7228

15 1552 0.7⋅1764+0.3⋅1091.7228=1562.3169

16 1465 0.7⋅1552+0.3⋅1562.3169=1555.0951

17 1398 0.7⋅1465+0.3⋅1555.0951=1492.0285

18 1893 0.7⋅1398+0.3⋅1492.0285=1426.2086

19 1422 0.7⋅1893+0.3⋅1426.2086=1752.9626

20 2063 0.7⋅1422+0.3⋅1752.9626=1521.2888

21 1703 0.7⋅2063+0.3⋅1521.2888=1900.4866

22 1758 0.7⋅1703+0.3⋅1900.4866=1762.246

23 0.7⋅1758+0.3⋅1762.246=1759.2738

(1) (2) (3) (7)

(4) (5) (6)
yea Sale Exponential |%Error
Error |Error| Error2
r s Smoothing |

1 977 977

2 1127 977 1127-977=150 150 22500 13.31%

3 694 1082 694-1082=-388 388 150544 55.91%

4 1357 810.4 1357-810.4=546.6 546.6 298771.56 40.28%

5 1020 1193.02 1020-1193.02=-173.02 173.02 29935.9204 16.96%

6 1187 1071.906 1187-1071.906=115.094 115.094 13246.6288 9.7%

7 866 1152.4718 866-1152.4718=-286.4718 286.4718 82066.0922 33.08%

8 1459 951.9415 1459-951.9415=507.0585 507.0585 257108.2819 34.75%

1163-1306.8825=-
9 1163 1306.8825 143.8825 20702.1629 12.37%
143.8825
10 991 1206.1647 991-1206.1647=-215.1647 215.1647 46295.8647 21.71%

1411-
11 1411 1055.5494 355.4506 126345.1137 25.19%
1055.5494=355.4506

12 1323 1304.3648 1323-1304.3648=18.6352 18.6352 347.2697 1.41%

13 995 1317.4094 995-1317.4094=-322.4094 322.4094 103947.8521 32.4%

1764-
14 1764 1091.7228 672.2772 451956.5874 38.11%
1091.7228=672.2772

15 1552 1562.3169 1552-1562.3169=-10.3169 10.3169 106.4374 0.66%

16 1465 1555.0951 1465-1555.0951=-90.0951 90.0951 8117.119 6.15%

17 1398 1492.0285 1398-1492.0285=-94.0285 94.0285 8841.3619 6.73%

1893-
18 1893 1426.2086 466.7914 217894.2532 24.66%
1426.2086=466.7914

1422-1752.9626=-
19 1422 1752.9626 330.9626 109536.2204 23.27%
330.9626

2063-
20 2063 1521.2888 541.7112 293451.0568 26.26%
1521.2888=541.7112

1703-1900.4866=-
21 1703 1900.4866 197.4866 39000.9694 11.6%
197.4866

22 1758 1762.246 1758-1762.246=-4.246 4.246 18.0284 0.24%

5629.702 2280732.780
23 1759.2738 Total 434.76%
1 3

Forecasting errors

1. Mean absolute error (MAE), also called mean absolute deviation (MAD)
MAD=5629.7021 / 21=268.0811

2. Mean squared error (MSE)

MSE=2280732.7803 / 21=108606.3229
ALPHA = 0.9

(3)
(1) (2)
Exponential Smoothing
year Sales
(α=0.9)

1 977 977

2 1127 0.9⋅977+0.1⋅977=977

3 694 0.9⋅1127+0.1⋅977=1112

4 1357 0.9⋅694+0.1⋅1112=735.8

5 1020 0.9⋅1357+0.1⋅735.8=1294.88

6 1187 0.9⋅1020+0.1⋅1294.88=1047.488

7 866 0.9⋅1187+0.1⋅1047.488=1173.0488

8 1459 0.9⋅866+0.1⋅1173.0488=896.7049

9 1163 0.9⋅1459+0.1⋅896.7049=1402.7705

10 991 0.9⋅1163+0.1⋅1402.7705=1186.977

11 1411 0.9⋅991+0.1⋅1186.977=1010.5977

12 1323 0.9⋅1411+0.1⋅1010.5977=1370.9598

13 995 0.9⋅1323+0.1⋅1370.9598=1327.796

14 1764 0.9⋅995+0.1⋅1327.796=1028.2796

15 1552 0.9⋅1764+0.1⋅1028.2796=1690.428

16 1465 0.9⋅1552+0.1⋅1690.428=1565.8428

17 1398 0.9⋅1465+0.1⋅1565.8428=1475.0843

18 1893 0.9⋅1398+0.1⋅1475.0843=1405.7084

19 1422 0.9⋅1893+0.1⋅1405.7084=1844.2708

20 2063 0.9⋅1422+0.1⋅1844.2708=1464.2271

21 1703 0.9⋅2063+0.1⋅1464.2271=2003.1227

22 1758 0.9⋅1703+0.1⋅2003.1227=1733.0123
23 0.9⋅1758+0.1⋅1733.0123=1755.5012

(1) (2) (3) (7)

(4) (5) (6)
yea Sale Exponential |%Error
Error |Error| Error2
r s Smoothing |

1 977 977

2 1127 977 1127-977=150 150 22500 13.31%

3 694 1112 694-1112=-418 418 174724 60.23%

4 1357 735.8 1357-735.8=621.2 621.2 385889.44 45.78%

5 1020 1294.88 1020-1294.88=-274.88 274.88 75559.0144 26.95%

6 1187 1047.488 1187-1047.488=139.512 139.512 19463.5981 11.75%

7 866 1173.0488 866-1173.0488=-307.0488 307.0488 94278.9656 35.46%

8 1459 896.7049 1459-896.7049=562.2951 562.2951 316175.802 38.54%

1163-1402.7705=-
9 1163 1402.7705 239.7705 57489.8869 20.62%
239.7705

10 991 1186.977 991-1186.977=-195.977 195.977 38407.0037 19.78%

1411-
11 1411 1010.5977 400.4023 160321.9979 28.38%
1010.5977=400.4023

12 1323 1370.9598 1323-1370.9598=-47.9598 47.9598 2300.1396 3.63%

13 995 1327.796 995-1327.796=-332.796 332.796 110753.1623 33.45%

1764-
14 1764 1028.2796 735.7204 541284.5104 41.71%
1028.2796=735.7204

15 1552 1690.428 1552-1690.428=-138.428 138.428 19162.3 8.92%

1465-1565.8428=-
16 1465 1565.8428 100.8428 10169.2695 6.88%
100.8428

17 1398 1475.0843 1398-1475.0843=-77.0843 77.0843 5941.9862 5.51%

 1893-
18 1893 1405.7084 487.2916 237453.0762 25.74%
1405.7084=487.2916

1422-1844.2708=-
19 1422 1844.2708 422.2708 178312.6647 29.7%
422.2708

2063-
20 2063 1464.2271 598.7729 358529.0046 29.02%
1464.2271=598.7729

1703-2003.1227=-
21 1703 2003.1227 300.1227 90073.6401 17.62%
300.1227

22 1758 1733.0123 1758-1733.0123=24.9877 24.9877 624.3866 1.42%

6575.362 2899413.848
23 1755.5012 Total 504.39%
7 8

Forecasting errors

1. Mean absolute error (MAE), also called mean absolute deviation

(MAD)MAD=6575.3627/21=313.1125

2.Mean squared error (MSE)

MSE=2899413.8488/21=138067.3261

Now, let's create a chart in MS Excel to visualize the MAD and MSE values for each alpha. We will then
discuss which alpha is relatively better for the forecast based on these metrics.
The Revelation:

As we examine the outcomes, two alpha values stand out as strong contenders: 0.2 and 0.5. They
reveal the lowest MAD and MSE values, signifying their prowess in forecasting accuracy. These alphas
have tested themselves to be reliable allies in our quest for precision.

The Concluding Remark:

Inside the grand forecasting arena, where accuracy is paramount, alpha values of 0.2 and 0.5 have
emerged as champions. Their capacity to limit errors, each absolute and squared, makes them the
desired selections for the ones looking for the most reliable forecasts.
So, in the long run, the selection of alpha is obvious: permit 0.2 and 0.5 to lead the way in our
forecasting endeavors, as they've proven themselves to be the keys to unlocking the doors of accurate
prediction.

Q3. A) You are not advised to use Software in this case, show the calculation-steps typed in your MS-
word file. In Roshni lights ( manufacturer of Mirchi lights) A production run of 2000 set of Mirchi lights
being tested for the life, the data shows that life is normally distributed with an average life span of 90
days and a standard deviation of 10 days. What is the probability if a set of Mirchi lights randomly
selected from that lot, survived up to 100 days?

B) Draw an appropriate chart to show the contribution of each category (in percent form) of Micro,
small and medium Enterprises (MSMEs) at the state level. (You may use the MS EXCEL). In conclusion
also write, why you have chosen this graph for this problem? (5 Marks)

Answer- To ascertain the chance of a randomly selected set of Mirchi lights from the production lot
surviving for 100 days, we delve into the world of records and regular distribution. With a median life
span (mean) of 90 days and a standard deviation of 10 days, we embark on a journey to calculate this
probability of the usage of the fascinating world of Z-scores and the venerable traditional typical
distribution desk.
Step 1: The Quest for the Z-Score

The Z-score, a numerical knight in shining armor, is our manual for understanding how far a particular
value ventures from the mean regarding general deviations. On this quest, we searched for the Z-score
for a Mirchi mild's endurance of 100 days. The Z-score reveals itself through this incantation:

𝑋−µ
𝑍=
𝜎

Where:

• X, the valorous value of interest, stands at 100 days.

• µ, the noble mean (average life span), holds steadfast at 90 days.
• σ, the trusty standard deviation, is a gallant 10 days

After invoking this magical formula, the Z-score emerges:

100 − 90
𝑍= =1
2

Step 2: The Revelation of Probability

With the Z-score at our side, we project to discover the opportunity concealed inside the annals of a
typical general distribution desk. This table, corresponding to an ancient manuscript, unravels the
secrets of probabilities associated with Z-scores.

Our quest is to unveil the probability that the Z-score is not only much less than one but additionally
identical to or under it (P (Z ≤ 1)). The standard ordinary distribution table, our mystical oracle, famous
that P (Z ≤ 1) is approximately 0.8413.

Thus, the revelation is clear: an enchanting probability of about 0.8413 exists, which translates to
84.13%. That is the opportunity for a randomly chosen set of Mirchi lights from the lot to endure
gallantly for 100 days.

In summary, the journey through the enigmatic world of statistics unveils an approximate 84.13%
chance. It foretells that a randomly selected assembly of Mirchi lighting from the production lot shall
withstand the take-a-look-at time and gleam for 100 days, guided by the mean and preferred deviation
of their life, and with the assumption of a normal distribution.
SOLUTION 3-B

Stacked Bar
14000

12000

10000

8000

6000

4000

2000

Micro Small Medium

Reasons for Choosing a Stacked Bar Chart:

Comparative Clarity: The number one objective is to offer a clear, visible assessment of the
contributions made by Micro, Small, and Medium enterprises to the overall range of MSMEs in each
district of Uttarakhand. A stacked bar chart excels on this, presenting those contributions facet
through side within each section. This visible contrast is crucial for comprehending the distribution of
MSME classes across the state.

Part-to-complete Visualization: Stacked bar charts inherently constitute the part-to-whole

relationship. Each bar in the diagram symbolizes a district's total range of MSMEs. Within these bars,
segments represent the 3 MSME classes - Micro, Small, and Medium. This department simplifies how
every type contributes to the overall number of MSMEs in a particular district. It effectively breaks
down the full into its constituent additives.

District-clever Insights: Uttarakhand is a country with various financial sports throughout its districts.
The stacked bar chart facilitates a district-wise contrast of MSME composition. By examining the
proportions of Micro, Small, and Medium corporations in each section, stakeholders can benefit from
insights into nearby versions of the types of businesses. Based on this evaluation, Policymakers and
strategists can tailor their strategies to different districts.

Clarity and Accessibility: Stacked bar charts are famous for their clarity and ease of interpretation.
They're intuitive and require minimal explanation. Each district is represented as a single bar, and
segments in the bars are frequently shade-coded for enhanced clarity. This design ensures that the
chart is out there and understandable to a broad target market, irrespective of their familiarity with
information visualization strategies. The visible distinction through coloration coding significantly
improves readability, enabling viewers to differentiate between MSME categories effortlessly.

In conclusion:

The Decision of a stacked bar chart as the desired visualization method for showing the contribution
of Micro, Small, and Medium organizations to the MSMEs in Uttarakhand is underpinned by its
suitability in numerous critical elements. It affords a clear basis for comparative analysis, effectively
communicates the part-to-whole relationship, facilitates district-wise comparisons, and excels in
phrases of clarity and accessibility. This decision ensures that the facts are presented in an informative
and understandable manner to a vast audience, making it a super tool for conveying insights into the
distribution of MSME categories within the state.