Exponential Smoothing Tool
The Exponential Smoothing tool performs the exponential smoothing for the given set or sets of values. Each value in the smoothed set is predicted based on the forecast for the prior period. The formula to calculate the forecast is: F(t+1) = F(t) + (1 - dampingFactor) * (A(t) - F(t)), where A(t) is the tth value in the original data set.
Specify the cells containing the datasets in the “Input Range” entry. The entered range or ranges are grouped into datasets either by rows or by columns.
If you have labels in the first cell of each data set, select the “Labels” option.
Specify prior forecast adjustment value in the “Damping factor” entry. A value, for example, between 0.2 and 0.3 represents 20 to 30 percent error adjustment in the prior forecast.
If you want to have the standard errors output as well, press the checkbutton on before starting the tool. The standard errors are calculated using the following formula: e(t) = SQRT ( ((A(t-3)-F(t-3))^2 + (A(t-2)-F(t-2))^2 + (A(t-1)-F(t-1))^2) / 3 ).
Figure 6-26 shows some example data and Figure 6-27 the corresponding output.
