Determination the smoothing constant that minimizes mean absolute error and mean square deviation
Exponential smoothing technique has become one of the quantitative techniques are very important in forecasting. The accuracy of forecasting based on this method depends on a parameter called the smoothing constant. Selection of smoothing constant value becomes very crucial because in forecasting...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2021
|
| Subjects: | |
| Online Access: | http://eprints.unisza.edu.my/4163/1/FH03-FIK-21-56567.pdf http://eprints.unisza.edu.my/4163/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Exponential smoothing technique has become one of the quantitative techniques are very important in
forecasting. The accuracy of forecasting based on this method depends on a parameter called the
smoothing constant. Selection of smoothing constant value becomes very crucial because in
forecasting prosecuted forecasting error is minimal. This paper discusses the selection of the optimal
smoothing constant value which minimizes the mean square error (MSE) and the mean absolute
deviation (MAD). Trial and error method is used to determine the optimal value of the smoothing
constant based on the two criterias (MSE and MAD). Based on the analysis carried out, there was no
regularity of the relationship between the amount of data and the smoothing constant value that
minimized MAD and MSE. |
|---|