Scribd
Upload
2K views3 pages

Projects Prasanna Chandra 7E Ch4 Minicase Solution

The document discusses three different forecasting methods: 1) Linear regression is used to estimate the parameters a and b in the linear relationship Yt = a + bT. This results in the linear regression Y = 507.89 + 254.93T. 2) Exponential smoothing is used to forecast values for periods 12-15 with a smoothing parameter of 0.4. 3) The moving average method forecasts values for periods 12-15 by taking the average of the last 4 periods.

Uploaded by

Jeet
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2K views3 pages

Projects Prasanna Chandra 7E Ch4 Minicase Solution

The document discusses three different forecasting methods: 1) Linear regression is used to estimate the parameters a and b in the linear relationship Yt = a + bT. This results in the linear regression Y = 507.89 + 254.93T. 2) Exponential smoothing is used to forecast values for periods 12-15 with a smoothing parameter of 0.4. 3) The moving average method forecasts values for periods 12-15 by taking the average of the last 4 periods.

Uploaded by

Jeet
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

4.

(i) We have to estimate the parameters a and b in the linear relationship


Yt = a + bT
Using the least squares method.
According to the least squares method the parameters are:

∑TY–nTY
b=
∑T2–nT2

a = Y – bT
The parameters are calculated below:
Calculations in the Least Squares Method
T Y TY T2
1 1,000 1,000 1
2 1,150 2,300 4
3 1,320 3,960 9
4 1,600 6,400 16
5 1,750 8,750 25
6 2,050 12,300 36
7 1,900 13,300 49
8 2,400 19,200 64
9 2,650 23,850 81
10 3,040 30,400 100
11 2,950 32,450 121
12 3,500 42,000 144
13 4,050 52,650 169
14 4,250 59,500 196
15 4,600 69,000 225
∑ T = 120 ∑ Y = 38,210 ∑ TY = 377,060 ∑ T = 1,240
2

T=8 Y = 2,547.33

∑TY–nTY 377,060 – 15 x 8 x 2,547.33


b= =
∑T2–nT2 1,240 – 15 x 8 x 8

71,380.4
= = 254.93
280
a = Y – bT
= 2,547.33 – 254.93 (8)
= 507.89
Thus, linear regression is
Y = 507.89 + 254.93 T
(ii) In general, in exponential smoothing the forecast for t + 1 is
Ft + 1 = Ft + α et

Where Ft + 1 = forecast for year t+1 α = smoothing parameter


et = error in the forecast for year t = St - Ft
F11 is given to be 3,250 and α is given to be 0.4
The forecasts for periods 12 to 15 are calculated below:

Period Data (St) Forecast Error Forecast for t + 1


t (Ft) (et = St -Ft) (Ft + 1 = Ft + 1 + α et)

1 1,000
2 1,150
3 1,320
4 1,600
5 1,750
6 2,050
7 1,900
8 2,400
9 2,650
10 3,040
11 2,950 3,250 -300 F12 = 3,250 + 0.4(-300) = 3,130
12 3,500 3,130 370 F13 = 3,130 + 0.4(370) = 3,278
13 4,050 3,278 772 F14 = 3,278 + 0.4(772) = 3,586.8
14 4,250 3,586.8 663.2 F15 = 3,586.8 + 0.4(663.2) = 3,852.08
15 4,600 3,852.08 747.92 F16 = 3,852.08 + 0.4(747.92) = 4,151.25
(iii) According to the moving average method
St + S t – 1 +…+ S t – n +1
Ft + 1 =
n
Where, Ft + 1 = forecast for the next period
St = sales for the current period
n = period over which averaging is done

Given n = 4, the forecasts for the periods 12 to 15 are given below:

Period Data (St) Forecast Forecast for t + 1


t (Ft) Ft + 1 = (St + S t – 1 + S t – 2 + S t – 3) / 4

1 1,000
2 1,150
3 1,320
4 1,600
5 1,750
6 2,050
7 1,900
8 2,400
9 2,650
10 3,040
11 2,950 F12 = (2,400 + 2,650 + 3,040 + 2,950)/ 4 = 2,760
12 3,500 2,760 F13 = (2,650 + 3,040 + 2,950 + 3,500)/ 4 = 3,035
13 4,050 3,035 F14 = (3,040 + 2,950 + 3,500 + 4,050) /4 = 3,385
14 4,250 3,385 F15 = (2,950 + 3,500 + 4,050 + 4,250) /4 = 3,687.5
15 4,600 3,687.5

You might also like