Forecasting Techniques Using Moving Average, Exponential Smoothing, and Weighted Moving Average
Forecasting is an attempt to predict the future using either quantitative or qualitative technique. Forecasting is an integral part of human activity, however, businesses are increasingly using the forecasting technique to predict sales, demand planning, cost projection, inventory control, corporate planning, advertising planning, production planning and investment cash flow. (Lucey, 2002). While there are different strategies that can be used for the forecasting, however, the time-series analysis is one of the effective strategies that businesses use for the forecasting. The time series analysis is a form of the statistical or mathematical technique using the past data to forecast the future. The benefit of the time series analysis is the simplicity. The examples of the time series analysis are the moving average, weighted moving average and exponential smoothing.
The objective of the study is to use the moving average, weighted moving average and exponential smoothing to forecast the demand for the next three-quarter.
Moving Average Forecast
The study uses the data in Table 1 to calculate the moving average for the demand of the next three-quarter.
Table 1: Actual Demand Data
Quarter
Forecast
Actual Demand
4Q 2008
1Q 2009
2Q 2009
3Q 2009
4Q 2009
1Q 2010
2Q 2010
3Q 2010
4Q 2010
1Q 2011
Table 2: Three --Quarter Moving Average
Quarter
Forecast
Actual Demand
Error
Calculation
3-Quarter Moving Average
4Q 2008
1Q 2009
2Q 2009
3Q 2009
(220+215+210)/3
4Q 2009
(215+210+220)/3
1Q 2010
(210+220+225)/3
2Q 2010
(220+225+240)/3
3Q 2010
228,333
(225+240+255)/3
4Q 2010
231,111
(240+255.260)/3
1Q 2011
233,148
(255+260+270)/3
Exponential Smoothing Forecast
Quarter
Forecast
Actual Demand
Calculation
Forecast using Exponential Smoothing with Value 0.6
4Q 2008
(220+215+210+220+225+240)/6
1Q 2009
211.67+0.6 *(220-211.67)
2Q 2009
220.67+0.6 *(215-220.67)
3Q 2009
217.27+0,6*(210-217.27)
4Q 2009
212.91+0,6*(220-212.91)
1Q 2010
217.16+0,6*(225-217.16)
2Q 2010
221.87+0,6*(240-221.87)
3Q 2010
232.75+0,6*(255-232.75)
4Q 2010
246.10+0,6*(260-246.10)
1Q 2011
254.44+0,6*(270-254.44)
Weighted Moving Average
Quarter
Forecast
Actual Demand
Calculation
Forecasting with 3 WMA (0.50, .35, 0.15)
4Q 2008
1Q 2009
2Q 2009
3Q 2009
0.5*210+0.35*215+0.15*220
4Q 2009
0,5*220+0,35*210+0,15*215
1Q 2010
0,5*225+0,35*220+0,15*210
2Q 2010
0,5*240+0,35*225+0,15*220
3Q 2010
0,5*255+0,35*240+0,15*225
4Q 2010
=0,5*260+0,35*255+0,15*240
1Q 2011
Forecasting techniques used for 2008 and 2009 first two-quarters.
The forecasting technique used to solve to the forecast for the period is t= Actual demand for the period (t - 1)
Forecasting error with Weighted Average Technique
The forecasts are not 100% perfect, thus, it is essential to know the extent of relying on the forecast using the forecasting error .The formula below is used to measure the forecasting error:
Period
Quarter
Forecast
Actual Demand
Forecasts with Weighted Moving Average
Calculation
Forecasting
Error
1
4Q 2008
2
1Q 2009
3
2Q 2009
4
3Q 2009
220-213.25
6.75
5
4Q 2009
225-215.75
9.25
6
1Q 2010
240-221
19
7
2Q 2010
255-231.71
23.25
8
3Q 2010
260-245.25
14.75
9
4Q 2010
270-255.25
14.75
10
1Q 2011
213,25
MAD =
You’re 80% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.