Semester – 2 Internal Assignment

Subject: Decision Science

Answers

Ans. 1. Exponential smoothing is a forecasting technique for the forecast of time-series data that are stationary
or that include no significant trend, cyclical or seasonal effects. These techniques are called smoothing
techniques because they produce forecasts based on “smoothing out” the irregular fluctuation effects in the
time-series data.

Exponential smoothing is used to weight data from previous time periods with exponentially decreasing
importance in the forecast. It is calculated by multiplying the actual value for the present time period (Xt) by
a value between 0 and 1 (the exponential smoothing constant) referred to as α and adding that result to the
product of the present time period forecast (Ft) and (1-α). The essence of this procedure is that the new forecast
is a combination of the present forecast and the present actual value.

The formula for the exponential smoothing can be represented as below:

Ft+1 = α. Xt + (1-α). Ft

The value of α is determined by the forecaster. If α is taken less than 0.5, then the next forecast will be more
closer to the previous forecast than on the actual value. If α is taken more than 0.5, more weight is put on the
actual value than on the forecast value, hence, the next forecast value will be closer to the actual value.

The below table shows the Foreign Tourist Arrivals data and the calculations of the forecast values from 2002
onwards. Since for making the forecast we require previous time period data, and in the provided data we had
figures for year 1991 and 2001 to 2017. Thus, for the first two rows there is no forecast data calculation.

The calculations for forecast values, starts from year 2002 onwards wherein the forecast values are taken as
actual value of year 2001 (as per naïve forecasting model) which is 2.54 million.

From 2003 onwards, forecast values are calculated for various values of α = 0.2, 0.4, 0.6 and 0.8.

For example - In year 2003, forecast value for α = 0.2, 04, 0.6 and 0.8 will be calculated as below:

F2003 = α.X2002 + (1- α ).F2002 = 0.2*2.38 + (1-0.2)*2.54 = 0.2*2.38 + 0.8*2.54 = 2.51

F2003 = 0.4*2.38 + (1-0.4)*2.54 = 0.4*2.38 + 0.6*2.54 = 2.48

F2003 = 0.6*2.38 + (1-0.6)*2.54 = 0.6*2.38 + 0.4*2.54 = 2.44

 F2003 = 0.8*2.38 + (1-0.8)*2.54 = 0.8*2.38 + 0.2*2.54 = 2.41

Further, for calculation MAD Mean Absolute Deviation and MSE Mean Square Error, we have calculated
error of individual forecast (ei) for values of α = 0.2, 04, 0.6 and 0.8 in the below table.

Error of individual forecast is calculated by; ei = Xt - Ft

Similarly, MAD = Σ leil ÷ number of forecasts

And, MSE = Σ ei2 ÷ number of forecasts

Xt Ft ei leil ei^2
α= α= α= α= α= α= α= α= α= α= α= α= α= α= α= α=
Year FTAs 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8
1991 1.68 - - - - - - - - - - - - - - - -
2001 2.54 - - - - - - - - - - - - - - - -
2002 2.38 2.54 2.54 2.54 2.54 -0.16 -0.16 -0.16 -0.16 0.16 0.16 0.16 0.16 0.03 0.03 0.03 0.03
2003 2.73 2.51 2.48 2.44 2.41 0.22 0.25 0.29 0.32 0.22 0.25 0.29 0.32 0.05 0.06 0.08 0.10
2004 3.46 2.55 2.58 2.62 2.67 0.91 0.88 0.84 0.79 0.91 0.88 0.84 0.79 0.82 0.78 0.71 0.63
2005 3.92 2.73 2.93 3.12 3.30 1.19 0.99 0.80 0.62 1.19 0.99 0.80 0.62 1.41 0.98 0.64 0.38
2006 4.45 2.97 3.33 3.60 3.80 1.48 1.12 0.85 0.65 1.48 1.12 0.85 0.65 2.19 1.26 0.72 0.43
2007 5.08 3.27 3.78 4.11 4.32 1.81 1.30 0.97 0.76 1.81 1.30 0.97 0.76 3.29 1.70 0.94 0.58
2008 5.28 3.63 4.30 4.69 4.93 1.65 0.98 0.59 0.35 1.65 0.98 0.59 0.35 2.72 0.97 0.35 0.12
2009 5.17 3.96 4.69 5.04 5.21 1.21 0.48 0.13 -0.04 1.21 0.48 0.13 0.04 1.47 0.23 0.02 0.00
2010 5.78 4.20 4.88 5.12 5.18 1.58 0.90 0.66 0.60 1.58 0.90 0.66 0.60 2.49 0.81 0.44 0.36
2011 6.31 4.52 5.24 5.52 5.66 1.79 1.07 0.79 0.65 1.79 1.07 0.79 0.65 3.21 1.14 0.63 0.42
2012 6.58 4.88 5.67 5.99 6.18 1.70 0.91 0.59 0.40 1.70 0.91 0.59 0.40 2.90 0.83 0.35 0.16
2013 6.97 5.22 6.03 6.34 6.50 1.75 0.94 0.63 0.47 1.75 0.94 0.63 0.47 3.07 0.88 0.39 0.22
2014 7.68 5.57 6.41 6.72 6.88 2.11 1.27 0.96 0.80 2.11 1.27 0.96 0.80 4.46 1.62 0.92 0.65
2015 8.03 5.99 6.92 7.30 7.52 2.04 1.11 0.73 0.51 2.04 1.11 0.73 0.51 4.16 1.24 0.54 0.26
2016 8.8 6.40 7.36 7.74 7.93 2.40 1.44 1.06 0.87 2.40 1.44 1.06 0.87 5.77 2.07 1.13 0.76
2017 10.04 6.88 7.94 8.37 8.63 3.16 2.10 1.67 1.41 3.16 2.10 1.67 1.41 10.00 4.42 2.77 2.00
Total Σ 25.17 15.92 11.71 9.42 48.04 19.01 10.65 7.11

α = 0.2 α = 0.4 α = 0.6 α = 0.8

𝛴|𝑒𝑖 | 25.17 15.92 11.71 9.42
𝑀𝐴𝐷 =
𝑛 16 16 16 16
𝛴𝑒𝑖2 48.04 19.01 10.65 7.11
𝑀𝐴𝐷 =
𝑛 16 16 16 16

α = 0.2 α = 0.4 α = 0.6 α = 0.8

MAD 1.573 0.995 0.732 0.589
MSE 3.003 1.188 0.665 0.444
The values of MAD and MSE are lower for larger value of α = 0.8, i.e., MAD = 0.589 and MSE = 0.444. This
indicates that smaller the value of α, the measurement of error is huge whereas, when the value of α is larger the
measurement of error is low.

We can also plot a graph for Xt and Ft and verify the above observation by seeing forecast for different values of
α and which forecast is closer to the actual value.

Foreign Tourist Arrvials in India (in millions)

12

10

0
1991 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Xt Actual Ft α = 0.2 Ft α = 0.4 Ft α = 0.6 Ft α = 0.8

From above graph, it is observed that forecast value with α = 0.8 is closer to the actual values from 2002 to
2017 on the other hand forecast values with α = 0.2 is the farthest from the actual values. The smaller the
value of α, the lesser the impact of error has on new forecast and new forecast is more like the old forecast.

We can observe that the graph line of forecast values with α = 0.2 is moving with constant pace (in upwards
direction only), whereas the graph line of forecast values with α = 0.8 is moving (up and down movement)
like the actual values in the graph.

Further, below table shows the forecast figures of year 2018 calculated for different values of α = 0.2, 04,
0.6 and 0.8:

Xt Ft
Year FTAs α = 0.2 α = 0.4 α = 0.6 α = 0.8
2017 10.04 6.88 7.94 8.37 8.63
2018 7.51 8.78 9.37 9.76

The forecast values for 2018 year are 7.51, 8.78, 9.37 and 9.76 million for α = 0.2, 0.4, 0.6 and 0.8 respectively.
 Ans. 2. Correlation is a measure of the degree of relatedness of variables. The most well-known measure of
correlation is the Pearson product – moment coefficient of correlation, r. The value of r ranges from -1 to 0 to
+1. An r value of +1 indicates a perfect positive correlation meaning that as one variable increases in value,
the other variable also tends to increase. On the other hand, an r value of -1 indicates a perfect negative
correlation meaning that as one variable increases in value, the other variable tends to decrease. An r value of
0 means no linear relationship is present between the two variables (say x and y).

Pearson Product-Moment Correlation Coefficient (r) can be calculated by below formula:

𝛴𝑥𝛴𝑦
Σ(x − x̅)(y − y̅) 𝛴𝑥𝑦 −
𝑟= = 𝑛
√Σ(x − x̅ )2 (y − y̅ )2 2
√[𝛴𝑥 2 − (𝛴𝑥) ][𝛴𝑦 2 − (𝛴𝑦) ]
2
𝑛 𝑛

Where x and y are variables and n is the number of data values.

We have data for 32 nos. of states and union territories of India and variables a, b, c and d are allotted to the
4 types of data available. Total MSMEs will be denoted by (a), Active Companies by (b), In Migration of
Persons from other states as (c) and 2017-18 GSDP – Current Prices in crores as (d) for calculating correlation
between these variables.

(i) Calculation of correlations between In Migration of Persons from other states (c) and Total MSMEs
(a).

In Migration of
Total Persons from
State MSMEs other states a2 c2 a*c
(a) (Census of
India) (c )

ANDAMAN & NICOBAR

6,061 81,267 36,735,721 6,604,325,289 492,559,287
ISLANDS
ANDHRA PRADESH 641,929 1,591,890 412,072,841,041 2,534,113,772,100 1,021,880,355,810
ARUNACHAL PRADESH 60,932 136,010 3,712,708,624 18,498,720,100 8,287,361,320
ASSAM 20,189 495,699 407,595,721 245,717,498,601 10,007,667,111
BIHAR 902,520 1,111,954 814,542,350,400 1,236,441,698,116 1,003,560,724,080
CHANDIGARH 11,209 633,966 125,641,681 401,912,889,156 7,106,124,894
CHHATTISGARH 69,758 1,267,668 4,866,178,564 1,606,982,158,224 88,429,984,344
DELHI 163,821 6,330,065 26,837,320,041 40,069,722,904,225 1,036,997,578,365
GOA 8,620 269,689 74,304,400 72,732,156,721 2,324,719,180
GUJARAT 807,378 3,916,075 651,859,234,884 15,335,643,405,625 3,161,752,801,350
HARYANA 185,486 3,626,318 34,405,056,196 13,150,182,237,124 672,631,220,548
HIMACHAL PRADESH 14,674 395,504 215,326,276 156,423,414,016 5,803,625,696
JAMMU & KASHMIR 9,399 155,187 88,341,201 24,083,004,969 1,458,602,613
JHARKHAND 157,813 2,195,521 24,904,942,969 4,820,312,461,441 346,481,755,573
 KARNATAKA 332,872 3,247,660 110,803,768,384 10,547,295,475,600 1,081,055,079,520
KERALA 125,934 654,423 15,859,372,356 428,269,462,929 82,414,106,082
MADHYA PRADESH 934,293 2,744,332 872,903,409,849 7,531,358,126,224 2,564,010,177,276
MAHARASHTRA 1,692,859 9,087,380 2,865,771,593,881 82,580,475,264,400 15,383,653,019,420
MANIPUR 34,423 20,100 1,184,942,929 404,010,000 691,902,300
MEGHALAYA 2,208 107,915 4,875,264 11,645,647,225 238,276,320
MIZORAM 3,238 41,380 10,484,644 1,712,304,400 133,988,440
NAGALAND 1,543 108,020 2,380,849 11,668,320,400 166,674,860
ODISHA 119,291 855,096 14,230,342,681 731,189,169,216 102,005,256,936
PUDUCHERRY 10,539 339,967 111,070,521 115,577,561,089 3,582,912,213
PUNJAB 203,394 2,488,299 41,369,119,236 6,191,631,913,401 506,105,086,806
RAJASTHAN 572,546 2,604,298 327,808,922,116 6,782,368,072,804 1,491,080,402,708
SIKKIM 875 61,163 765,625 3,740,912,569 53,517,625
TAMIL NADU 1,032,490 1,650,771 1,066,035,600,100 2,725,044,894,441 1,704,404,549,790
TRIPURA 5,936 87,378 35,236,096 7,634,914,884 518,675,808
UTTAR PRADESH 887,413 4,061,933 787,501,832,569 16,499,299,696,489 3,604,612,149,329
UTTARAKHAND 40,443 1,250,575 1,635,636,249 1,563,937,830,625 50,577,004,725
WEST BENGAL 231,190 2,381,045 53,448,816,100 5,669,375,292,025 550,473,793,550
Total 9,291,276 53,998,548 8,132,866,747,168 221,081,999,514,428 34,492,991,653,879

𝛴𝑎𝛴𝑐
𝛴𝑎𝑐− 𝑛
𝑟𝑎𝑐 = 2 2
√[𝛴𝑎2 −(𝛴𝑎) ][𝛴𝑐 2 −(𝛴𝑐) ]
𝑛 𝑛

(𝟗,𝟐𝟗𝟏,𝟐𝟕𝟔)∗(𝟓𝟑,𝟗𝟗𝟖,𝟓𝟒𝟖)
𝟑𝟒,𝟒𝟗𝟐,𝟗𝟗𝟏,𝟔𝟓𝟑,𝟖𝟕𝟗 − 32
= = 0.708
2 2
√[𝟖,𝟏𝟑𝟐,𝟖𝟔𝟔,𝟕𝟒𝟕,𝟏𝟔𝟖 − (𝟗,𝟐𝟗𝟏,𝟐𝟕𝟔) ][𝟐𝟐𝟏,𝟎𝟖𝟏,𝟗𝟗𝟗,𝟓𝟏𝟒,𝟒𝟐𝟖 − (𝟓𝟑,𝟗𝟗𝟖,𝟓𝟒𝟖) ]
32 32

The value of rac is 0.708 which means that there is moderate positive correlation between the in migration of
persons and the total number of MSMEs. We can conclude that the states with higher number of MSMEs created
more job opportunities and thus more number of people migrated to these states.

(ii) Calculation of correlations between In Migration of Persons from other states (c) and Active
companies (b).

In Migration of
Active Persons from
State Companies other states b2 c2 b*c
(b) (Census of India)
(c )
ANDAMAN &
319 81,267 101,761 6,604,325,289 25,924,173
NICOBAR ISLANDS
ANDHRA PRADESH 116,210 1,591,890 13,504,764,100 2,534,113,772,100 184,993,536,900
ARUNACHAL
237 136,010 56,169 18,498,720,100 32,234,370
PRADESH
ASSAM 6,346 495,699 40,271,716 245,717,498,601 3,145,705,854
BIHAR 20,867 1,111,954 435,431,689 1,236,441,698,116 23,203,144,118
CHANDIGARH 7,956 633,966 63,297,936 401,912,889,156 5,043,833,496
CHHATTISGARH 7,109 1,267,668 50,537,881 1,606,982,158,224 9,011,851,812
DELHI 216,531 6,330,065 46,885,673,961 40,069,722,904,225 1,370,655,304,515
GOA 4,125 269,689 17,015,625 72,732,156,721 1,112,467,125
GUJARAT 62,619 3,916,075 3,921,139,161 15,335,643,405,625 245,220,700,425
HARYANA 30,868 3,626,318 952,833,424 13,150,182,237,124 111,937,184,024
HIMACHAL
3,344 395,504 11,182,336 156,423,414,016 1,322,565,376
PRADESH
JAMMU &
2,745 155,187 7,535,025 24,083,004,969 425,988,315
KASHMIR
JHARKHAND 9,554 2,195,521 91,278,916 4,820,312,461,441 20,976,007,634
KARNATAKA 68,333 3,247,660 4,669,398,889 10,547,295,475,600 221,922,350,780
KERALA 31,244 654,423 976,187,536 428,269,462,929 20,446,792,212
MADHYA
22,315 2,744,332 497,959,225 7,531,358,126,224 61,239,768,580
PRADESH
MAHARASHTRA 231,912 9,087,380 53,783,175,744 82,580,475,264,400 2,107,472,470,560
MANIPUR 434 20,100 188,356 404,010,000 8,723,400
MEGHALAYA 560 107,915 313,600 11,645,647,225 60,432,400
MIZORAM 66 41,380 4,356 1,712,304,400 2,731,080
NAGALAND 235 108,020 55,225 11,668,320,400 25,384,700
ODISHA 15,204 855,096 231,161,616 731,189,169,216 13,000,879,584
PUDUCHERRY 1,282 339,967 1,643,524 115,577,561,089 435,837,694
PUNJAB 16,909 2,488,299 285,914,281 6,191,631,913,401 42,074,647,791
RAJASTHAN 37,022 2,604,298 1,370,628,484 6,782,368,072,804 96,416,320,556
SIKKIM 2 61,163 4 3,740,912,569 122,326
TAMIL NADU 76,675 1,650,771 5,879,055,625 2,725,044,894,441 126,572,866,425
TRIPURA 317 87,378 100,489 7,634,914,884 27,698,826
UTTAR PRADESH 70,863 4,061,933 5,021,564,769 16,499,299,696,489 287,840,758,179
UTTARAKHAND 4,792 1,250,575 22,963,264 1,563,937,830,625 5,992,755,400
WEST BENGAL 135,844 2,381,045 18,453,592,336 5,669,375,292,025 323,450,676,980
Total 1,202,839 53,998,548 157,175,027,023 221,081,999,514,428 5,284,097,665,610

𝛴𝑏𝛴𝑐
𝛴𝑏𝑐− 𝑛
𝑟𝑏𝑐 = 2 2
√[𝛴𝑏2 −(𝛴𝑏) ][𝛴𝑐 2 −(𝛴𝑐) ]
𝑛 𝑛

(𝟏,𝟐𝟎𝟐,𝟖𝟑𝟗)∗(𝟓𝟑,𝟗𝟗𝟖,𝟓𝟒𝟖)
𝟓,𝟐𝟖𝟒,𝟎𝟗𝟕,𝟔𝟔𝟓,𝟔𝟏𝟎 − 32
= = 0.853
2 2
√[𝟏𝟓𝟕,𝟏𝟕𝟓,𝟎𝟐𝟕,𝟎𝟐𝟑 − (𝟏,𝟐𝟎𝟐,𝟖𝟑𝟗) ][𝟐𝟐𝟏,𝟎𝟖𝟏,𝟗𝟗𝟗,𝟓𝟏𝟒,𝟒𝟐𝟖 − (𝟓𝟑,𝟗𝟗𝟖,𝟓𝟒𝟖) ]
32 32
The value of rbc is 0.853 which means that there is strong positive correlation between the in migration of persons
and the active companies. It indicates that the states with higher number of active companies created more growth
and job opportunities. Hence, more number of people migrated into these states for better livelihood. And the
states with lesser active companies saw less migration of people.

(iii) Calculation of correlations between In Migration of Persons from other states (c) and 2017-18 GSDP
– Current Prices (in Crores) (d).

In Migration of 2017-18
Persons from GSDP -
State other states CURRENT c2 d2 c*d
(Census of India) PRICES (` in
(c) Crore) (d)

ANDAMAN & NICOBAR

81,267 7,871 6,604,325,289 61,952,641 639,652,557
ISLANDS
ANDHRA PRADESH 1,591,890 1,546,313 2,534,113,772,100 2,391,083,893,969 2,461,560,201,570
ARUNACHAL PRADESH 136,010 22,432 18,498,720,100 503,194,624 3,050,976,320
ASSAM 495,699 288,691 245,717,498,601 83,342,493,481 143,103,840,009
BIHAR 1,111,954 484,740 1,236,441,698,116 234,972,867,600 539,008,581,960
CHANDIGARH 633,966 38,760 401,912,889,156 1,502,337,600 24,572,522,160
CHHATTISGARH 1,267,668 484,740 1,606,982,158,224 234,972,867,600 614,489,386,320
DELHI 6,330,065 686,824 40,069,722,904,225 471,727,206,976 4,347,640,563,560
GOA 269,689 70,494 72,732,156,721 4,969,404,036 19,011,456,366
GUJARAT 3,916,075 1,328,068 15,335,643,405,625 1,763,764,612,624 5,200,813,893,100
HARYANA 3,626,318 649,592 13,150,182,237,124 421,969,766,464 2,355,627,162,256
HIMACHAL PRADESH 395,504 138,351 156,423,414,016 19,140,999,201 54,718,373,904
JAMMU & KASHMIR 155,187 137,427 24,083,004,969 18,886,180,329 21,326,883,849
JHARKHAND 2,195,521 276,243 4,820,312,461,441 76,310,195,049 606,497,307,603
KARNATAKA 3,247,660 1,357,579 10,547,295,475,600 1,843,020,741,241 4,408,955,015,140
KERALA 654,423 701,577 428,269,462,929 492,210,286,929 459,128,125,071
MADHYA PRADESH 2,744,332 724,729 7,531,358,126,224 525,232,123,441 1,988,896,986,028
MAHARASHTRA 9,087,380 2,411,600 82,580,475,264,400 5,815,814,560,000 21,915,125,608,000
MANIPUR 20,100 23,835 404,010,000 568,107,225 479,083,500
MEGHALAYA 107,915 30,790 11,645,647,225 948,024,100 3,322,702,850
MIZORAM 41,380 18,737 1,712,304,400 351,075,169 775,337,060
NAGALAND 108,020 24,492 11,668,320,400 599,858,064 2,645,625,840
ODISHA 855,096 434,769 731,189,169,216 189,024,083,361 371,769,232,824
PUDUCHERRY 339,967 32,962 115,577,561,089 1,086,493,444 11,205,992,254
PUNJAB 2,488,299 478,636 6,191,631,913,401 229,092,420,496 1,190,989,480,164
RAJASTHAN 2,604,298 835,170 6,782,368,072,804 697,508,928,900 2,175,031,560,660
SIKKIM 61,163 23,495 3,740,912,569 552,015,025 1,437,024,685
TAMIL NADU 1,650,771 1,461,841 2,725,044,894,441 2,136,979,109,281 2,413,164,729,411
TRIPURA 87,378 44,219 7,634,914,884 1,955,319,961 3,863,767,782
UTTAR PRADESH 4,061,933 1,460,443 16,499,299,696,489 2,132,893,756,249 5,932,221,616,319
 UTTARAKHAND 1,250,575 222,836 1,563,937,830,625 49,655,882,896 278,673,130,700
WEST BENGAL 2,381,045 999,585 5,669,375,292,025 999,170,172,225 2,380,056,866,325
Total 53,998,548 17,447,841 221,081,999,514,428 20,839,870,930,201 59,929,802,686,147

𝛴𝑐𝛴𝑑
𝛴𝑐𝑑 −
𝑟𝑐𝑑 = 𝑛
2 2
√[𝛴𝑐 2 − (𝛴𝑐) ][𝛴𝑑2 − (𝛴𝑑) ]
𝑛 𝑛

(𝟓𝟑,𝟗𝟗𝟖,𝟓𝟒𝟖)∗(𝟏𝟕,𝟒𝟒𝟕,𝟖𝟒𝟏)
𝟓𝟗,𝟗𝟐𝟗,𝟖𝟎𝟐,𝟔𝟖𝟔,𝟏𝟒𝟕 − 32
= = 0.795
(𝟓𝟑,𝟗𝟗𝟖,𝟓𝟒𝟖) 2 (𝟏𝟕,𝟒𝟒𝟕,𝟖𝟒𝟏)2
√[𝟐𝟐𝟏,𝟎𝟖𝟏,𝟗𝟗𝟗,𝟓𝟏𝟒,𝟒𝟐𝟖 − ][𝟐𝟎,𝟖𝟑𝟗,𝟖𝟕𝟎,𝟗𝟑𝟎,𝟐𝟎𝟏 − ]
32 32

The above value of rcd of 0.795 indicates strong positive correlation between the Gross State Domestic Product
(GSDP) and the in migration of people from other states. More people migrate to the developing states in search
of better opportunities and standard of living. The better the GSDP more number of people migrate to these states.
The GSDP is a measure in monetary terms, the sum total volume of all finished goods and services produced
during a given period of time. Large number of MSMEs and active companies, create more products and services
for the people, organizations and thus contribute to the GSDP, also these companies grow and create job
opportunities for people who migrate to these states. The migrated people also contribute to the GSDP by not
only producing the goods and services but also consuming some amount of goods and services for themselves.

(iv) Calculation of correlations between Total MSMEs (a) and Active companies (b).

Total Active
State a2 b2 a*b
MSMEs (a) Companies (b)
ANDAMAN & NICOBAR
6,061 319 36,735,721 101,761 1,933,459
ISLANDS
ANDHRA PRADESH 641,929 116,210 412,072,841,041 13,504,764,100 74,598,569,090
ARUNACHAL PRADESH 60,932 237 3,712,708,624 56,169 14,440,884
ASSAM 20,189 6,346 407,595,721 40,271,716 128,119,394
BIHAR 902,520 20,867 814,542,350,400 435,431,689 18,832,884,840
CHANDIGARH 11,209 7,956 125,641,681 63,297,936 89,178,804
CHHATTISGARH 69,758 7,109 4,866,178,564 50,537,881 495,909,622
DELHI 163,821 216,531 26,837,320,041 46,885,673,961 35,472,324,951
GOA 8,620 4,125 74,304,400 17,015,625 35,557,500
GUJARAT 807,378 62,619 651,859,234,884 3,921,139,161 50,557,202,982
HARYANA 185,486 30,868 34,405,056,196 952,833,424 5,725,581,848
HIMACHAL PRADESH 14,674 3,344 215,326,276 11,182,336 49,069,856
JAMMU & KASHMIR 9,399 2,745 88,341,201 7,535,025 25,800,255
JHARKHAND 157,813 9,554 24,904,942,969 91,278,916 1,507,745,402
KARNATAKA 332,872 68,333 110,803,768,384 4,669,398,889 22,746,142,376
 KERALA 125,934 31,244 15,859,372,356 976,187,536 3,934,681,896
MADHYA PRADESH 934,293 22,315 872,903,409,849 497,959,225 20,848,748,295
MAHARASHTRA 1,692,859 231,912 2,865,771,593,881 53,783,175,744 392,594,316,408
MANIPUR 34,423 434 1,184,942,929 188,356 14,939,582
MEGHALAYA 2,208 560 4,875,264 313,600 1,236,480
MIZORAM 3,238 66 10,484,644 4,356 213,708
NAGALAND 1,543 235 2,380,849 55,225 362,605
ODISHA 119,291 15,204 14,230,342,681 231,161,616 1,813,700,364
PUDUCHERRY 10,539 1,282 111,070,521 1,643,524 13,510,998
PUNJAB 203,394 16,909 41,369,119,236 285,914,281 3,439,189,146
RAJASTHAN 572,546 37,022 327,808,922,116 1,370,628,484 21,196,798,012
SIKKIM 875 2 765,625 4 1,750
TAMIL NADU 1,032,490 76,675 1,066,035,600,100 5,879,055,625 79,166,170,750
TRIPURA 5,936 317 35,236,096 100,489 1,881,712
UTTAR PRADESH 887,413 70,863 787,501,832,569 5,021,564,769 62,884,747,419
UTTARAKHAND 40,443 4,792 1,635,636,249 22,963,264 193,802,856
WEST BENGAL 231,190 135,844 53,448,816,100 18,453,592,336 31,405,774,360
Total 9,291,276 1,202,839 8,132,866,747,168 157,175,027,023 827,790,537,604

𝛴𝑎𝛴𝑏
𝛴𝑎𝑏−
𝑛
𝑟𝑎𝑏 = 2 2
√[𝛴𝑎2 −(𝛴𝑎) ][𝛴𝑏2 −(𝛴𝑏) ]
𝑛 𝑛

(𝟗,𝟐𝟗𝟏,𝟐𝟕𝟔)∗(𝟏,𝟐𝟎𝟐,𝟖𝟑𝟗)
𝟖𝟐𝟕,𝟕𝟗𝟎,𝟓𝟑𝟕,𝟔𝟎𝟒 −
32
= = 0.613
(𝟗,𝟐𝟗𝟏,𝟐𝟕𝟔)2 (𝟏,𝟐𝟎𝟐,𝟖𝟑𝟗)2
√[𝟖,𝟏𝟑𝟐,𝟖𝟔𝟔,𝟕𝟒𝟕,𝟏𝟔𝟖 − ][𝟏𝟓𝟕,𝟏𝟕𝟓,𝟎𝟐𝟕,𝟎𝟐𝟑 − ]
32 32

The value of rab is 0.613 which shows that there is moderate positive correlation between total MSMEs and
active companies. The state with higher number of MSMEs also have high number of active companies.

Ans. 3 a. Mean and Standard Deviation:

Mean: The mean (average) of a data set is found by adding all numbers in the data set and then dividing by the
number of values in the set. It can be denoted by μ.

Standard Deviation: Standard Deviation is a measure, which shows how much variation (such as spread,
dispersion) from the mean exists. The standard deviation indicates a “typical” deviation from the mean. It is a
popular measure of variability because it returns to the original units of measure of the data set. Like the variance,
if the data points are close to the mean, there is a small variation whereas the data points are highly spread out
from the mean, then it has a high variance. Standard deviation calculates the extent to which the values differ
 from the average. Standard Deviation is the most widely used measure of dispersion, and is based on all values.
Therefore, a change in even one value affects the value of standard deviation. It is independent of origin but not
of scale. It is also useful in certain advanced statistical problems. It is denoted by σ.

Number Of
Indigenous Total
S.
No.
State District (Desi) Buffalo (x-x̅ ) (y-y̅) (x-x̅ )2 (y-y̅)2
Total (y)
Cattle (x)
1 UTTARAKHAND Uttarkashi 82991 25945 -15109 -40695 228288854.44 1656070503.49
2 UTTARAKHAND Chamoli 132165 37922 34065 -28718 1160408502.75 824714687.72
3 UTTARAKHAND Rudraprayag 71277 31115 -26823 -35525 719485708.90 1262014694.25
4 UTTARAKHAND Tehri Garhwal 80833 79394 -17267 12754 298157258.44 162668440.33
5 UTTARAKHAND Dehradun 87753 52185 -10347 -14455 107065184.59 208942577.33
6 UTTARAKHAND Garhwal 235412 30076 137312 -36564 18854521969.28 1336914845.56
7 UTTARAKHAND Pithoragarh 108417 37056 10317 -29584 106435727.36 875203953.25
8 UTTARAKHAND Bageshwar 70863 31432 -27237 -35208 741866739.98 1239592430.79
9 UTTARAKHAND Almora 132038 77444 33938 10804 1151772180.36 116729740.33
10 UTTARAKHAND Champawat 45506 18599 -52594 -48041 2766153110.21 2307922899.18
11 UTTARAKHAND Nainital 101009 77759 2909 11119 8460938.44 123635582.25
12 UTTARAKHAND Udham Singh Nagar 63311 152911 -34789 86271 1210290577.51 7442711985.95
13 UTTARAKHAND Hardwar 63728 214480 -34372 147840 1181450248.05 21856711089.25
Total 1275303 866318 28534357000.31 39413833429.69
Mean = Total ÷ Number of data 98100 66640
(x−x̅)2
Standard Deviation of x = √
𝑁
Mean of x, x̅ = 98100 √(28534357000.31 ÷ 13) = 46850.30

̅)2
(y−y
Standard Deviation of y = √
𝑁
Mean of y, y̅ = 66640 √(39413833429.69 ÷ 13) = 55062.09

Number Of Indigenous (Desi) Total Cattle Total Buffalo

Distance from the Values within Values within
Mean Value Range Distance Value Range Distance
51250 11578
μ ± 1σ 92.31% 84.62%
144951 121702
4400 -43484
μ ± 2σ 92.31% 92.31%
191801 176764
-42451 -98546
μ ± 3σ 100.00% 100.00%
238651 231826

From above table, it is evident that 100% of the values of Indigenous cattle and buffalos are within 3 standard
deviations. This shows that the given data is normally distributed and also falls under empirical rule. The
empirical rule reveals the percentage of values that are within one, two or three standard deviations of the
mean for a set of data. A normal distribution is symmetric from the peak of the curve, where the mean is. This
means that most of the observed data is clustered near the mean, while the data become less frequent when
farther away from the mean.

Below I have calculated normal distribution values of the data given and plotted its graph to see the
distribution of values. Both the graphs seem to be similar in nature. Most of the values are clustered near the
mean and only few values are at the far end of the graph. The graph is going up and slowing coming down.

The above graphs also reminds of exponential distribution as the graphs are slightly skewed to the right.

Ans. 3 b. Time Series Components: Time series data are data gathered on a given characteristic over a period
of time at regular intervals. For example in the question no. 3 b of the assignment, data of productivity of
Groundnut and, Rapeseed and Mustard in Kg per hectare for 71 years from 1950 to 2021 is provided. The
period of the data is given as per fiscal year, i.e., from 1st April to 31st March cycle.

Time series forecasting techniques attempt to account for changes over time by examining patterns, cycles,
or trends or using information about previous time periods to predict the outcome for future time period.
Generally, the time series data are composed of four elements: trend, cycles, seasonal effects and irregular
fluctuations. However, it is not necessary that all time-series data have all these elements.

The long term general direction of data is referred to as trend. In the below graph of 71 years we can see that
there is an upward trend in the productivity of both Groundnut and, Rapeseed and Mustard even though there
are numerous ups and downs in the graph.

Between years 1953-54 to 1964-65, the productivity of groundnut and the productivity of rapeseed and
mustard are stable and there are lesser fluctuations during these 11 years.
 Also, from 1950-51 to 1984-85, there is significant gap (represented by dotted lines in graph) in the
productivity of Groundnut and, Rapeseed and Mustard in Kg per hectare. And the productivity of Groundnut
is continuously more than the productivity of Rapeseed and Mustard during these years.

However, after 1985-86 the gap in the productivity of Groundnut and, Rapeseed and Mustard reduces. There
are also few instances (9 years) where the productivity of rapeseed and mustard have crossed the yield of
groundnut i.e., in year 1990-91, 1991-92, 1999-00, 2002-03, 2004-05, 2006-07, 2009-10, 2012-13 and 2018-
19.

The productivity of rapeseed and mustard has seen more growth from 1980-81 onwards than the productivity
of groundnut. In 1950-51 the productivity of rapeseed and mustard was at 368 Kg/ hectare which increased
to 1511 Kg/ hectare in the year 2022-21. On the other hand, the productivity of groundnut was at 775
Kg/hectare which has increased to 1676 Kg/ hectare in the year 2022-21 (highest recorded productivity of the
groundnut was 2063 Kg/ hectare in year 2019-20).

In the last 5-6 years data i.e., from 2015-16 to 2020-21, the productivity of groundnut and the productivity of
rapeseed and mustard is nearing each other at alternate years.

Productivity of Groundnut, Rapeseed and Mustard (Kg/Hectare)

(1950-2021)
2200
2000
PRODUCTION KG/HECTARE

1800
1600
1400
1200
1000
800
600
400
200
0
1958-59

1977-78

1984-85

1996-97

2003-04
1950-51
1951-52
1952-53
1953-54
1954-55
1955-56
1956-57
1957-58
1959-60
1960-61
1961-62
1962-63
1963-64
1964-65
1965-66
1966-67
1967-68
1968-69
1969-70
1970-71
1971-72
1972-73
1973-74
1974-75
1975-76
1976-77
1978-79
1979-80
1980-81
1981-82
1982-83
1983-84
1985-86
1986-87
1987-88
1988-89
1989-90
1990-91
1991-92
1992-93
1993-94
1994-95
1995-96
1997-98
1998-99
1999-00
2000-01
2001-02
2002-03
2004-05
2005-06
2006-07
2007-08
2008-09
2009-10
2010-11
2011-12
2012-13
2013-14
2014-15
2015-16
2016-17
2017-18
2018-19
2019-20
2020-21
YEARS

Productivity Groundnut (Kg/hectare) Productivity Rapeseed & Mustard (Kg./hectare)

Cycles are patterns of highs and lows through which data move over time periods usually of more than a year.
We can see in the below graph that the productivity of groundnut is showing high cyclic fluctuations from
year 1997-98 onwards. The productivity is increasing and decreasing regularly.
Similar cyclic fluctuations are there in productivity of rapeseed and mustard as the graph is moving upwards
and downwards. However, the fluctuations are slightly flat in comparison to the productivity of groundnut.
These cyclical effects can be plotted only if we have enough history data for a long period of time.

Further, seasonal effects, are shorter cycles, which usually occur in time periods of less than one year. These
effects are usually measured in months, quarter, weeks or even days. The above data does not depict any
seasonal effects.

Lastly, there are irregular fluctuations that are rapid changes in the data, which occur in even shorter time
frames than seasonal effects. The above data does not show any irregular fluctuations. Irregular fluctuations
can happen as often as day to day and they are subject to momentary change and are often unexplained. Time
series data that contain no trend, cyclical or seasonal fluctuations are said to be stationary. Techniques used
to forecast stationary data analyze only the irregular fluctuation effects.

**************************