0% found this document useful (0 votes)
46 views4 pages

GR7 Hw3mansci

1. The document discusses using different forecasting methods like simple moving averages, weighted moving averages, and exponential smoothing on time series data to calculate forecasts and forecast errors. It compares the mean squared errors of each method to determine the best forecasting approach. 2. For a second time series dataset on monthly sales, it calculates forecasts using 3-month simple moving averages and exponential smoothing. The 3-month moving average produces a lower mean squared error, indicating it provides a better forecast for this data. 3. Moving averages tend to smooth out fluctuations but cannot account for long-term trends or seasonality patterns in the data. They are better for short-term forecasts than just using the most recent observation.

Uploaded by

RIANE PADIERNOS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
46 views4 pages

GR7 Hw3mansci

1. The document discusses using different forecasting methods like simple moving averages, weighted moving averages, and exponential smoothing on time series data to calculate forecasts and forecast errors. It compares the mean squared errors of each method to determine the best forecasting approach. 2. For a second time series dataset on monthly sales, it calculates forecasts using 3-month simple moving averages and exponential smoothing. The 3-month moving average produces a lower mean squared error, indicating it provides a better forecast for this data. 3. Moving averages tend to smooth out fluctuations but cannot account for long-term trends or seasonality patterns in the data. They are better for short-term forecasts than just using the most recent observation.

Uploaded by

RIANE PADIERNOS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Group Assignment #3

Management Science: Moving Averages and Exponential Smoothing


1. Refer to page 769 item # 5 in the prescribed textbook (see item below):

Time Series Plot


20

15
Value

10

0
0 1 2 3 4 5 6 7
Week
a.
b.

Week Value Forecas Forecast Absolute Value of Squared


t Error Forecast Error Forecast Error
1 18
2 13
3 16
4 11 15.67 -4.67 4.67 21.81
5 17 13.33 3.67 3.67 13.47
6 14 14.67 -0.67 0.67 0.45
7 14
total -1.67 9.01 35.73

35.73
MSE = Forecast for Week 7 = 14
3
= 11.91
c.

Week Value Forecas Forecast Absolute Value Squared Forecast Error


t Error of Forecast
Error
1 18
2 13 18 -5 5 25
3 16 17 -1 1 1
4 11 16.8 -5.8 5.8 33.64
5 17 15.64 1.36 1.36 1.85
6 14 15.91 -1.91 1.91 3.65
7
total 83.35 -12.35 15.07 65.14

F3 = 0.2Y2 + 0.8F2 = 0.2(13) + 0.8(18) = 17


F4 = 0.2Y3 + 0.8F3 = 0.2(16) + 0.8(17) =16.8
F5 = 0.2Y4 + 0.8F4 = 0.2(11) + 0.8(16.8) =15.64
F6 = 0.2Y5 + 0.8F5 = 0.2(17) + 0.8(15.64) =15.91
F7 = 0.2Y6 + 0.8F6 = 0.2(14) + 0.8(15.91) = 15.53
65.14
MSE = Forecast for Week 7 = 15.53
5
= 13.03

d. The three-week moving average provides a better forecast because it has a smaller MSE.

Week Value Forecas Forecast Absolute Value of Squared Forecast


t Error Forecast Error Error
1 18
2 13 18 -5 5 25
3 16 15.5 0.5 0.5 0.25
4 11 15.75 -4.75 4.75 22.56
5 17 13.38 3.62 3.62 13.1
6 14 15.19 -1.19 1.19 1.42
7
total 77.82 -6.82 15.06 62.33
F3 = 0.5Y2 + 0.5F2 = 0.5(13) + 0.5(18) = 15.5
F4 = 0.5Y3 + 0.5F3 = 0.5(16) + 0.5(15.5) =15.75
F5 = 0.5Y4 + 0.5F4 = 0.5(11) + 0.5(15.75) =13.38
F6 = 0.5Y5 + 0.5F5 = 0.5(17) + 0.5(13.38) =15.19
F7 = 0.5Y6 + 0.5F6 = 0.5(14) + 0.5(15.19) = 14.60

6 2.33
e. MSE = =12.46
5
2. Refer to page 771 item #13 in the prescribed textbook (see item below):

400
350
300
millions of dollars

250
200
150
100
50
0
0 2 4 6 8 10 12 14
months
a.
b.

Week Value Forecas Forecast Absolute Value of Squared Forecast


t Error Forecast Error Error
1 240
2 350
3 230
4 260 273.33 -13.33 13.33 177.69
5 280 280 0 0 0
6 320 256.67 63.33 63.33 4010.69
7 220 286.67 -66.67 66.67 4444.89
8 310 273.33 36.67 36.67 1344.69
9 240 283.33 -43.33 43.33 1877.49
10 310 256.67 53.33 53.33 2844.09
11 240 286.67 -46.67 46.67 2178.09
12 230 263.33 -33.33 33.33 1110.89

MSE(3-Month) 17,988.52/9 = 1998.72


MSE (a = 0.2) = 27,818.49/11 = 2528.95
The 3-month moving average provides better forecast. Meanwhile, exponential smoothing
method is more difficult and complicated to forecast.
c. Using exponential smoothing,
F13 = a Y12 + (1 - a) F12 = 0.20(230) 0.80(267.53)
= 260 is the forecast for next month
3. What are the main problems of using moving averages in forecasting?
Moving average forecasts has tended to smooth out the random fluctuations in the time series.
However, the three-week moving average approach provided more accurate forecasts than
simply using the most recent observation as the forecast and using this method not quite useful or
cannot exhibits for long term picture of trend or seasonality.

You might also like