MOVING AVERAGES
| Collection | International Encyclopedia of Systems and Cybernetics |
|---|---|
| Year | 2004 |
| Vol. (num.) | 2(2) |
| ID | ◀ 2215 ▶ |
| Object type | Methodology or model |
- “A flexible mathematical method of smoothing statistical data ”(E. Dewey & E. Dakin, 1952, p. 248)
This smoothing process is of great interest for a better understanding of the fundamental evolving trends in growth curves.
The authors give the example of a 2 year moving average applied to a collection of annual data. For instance, it can be done in a progressive way by adding series of 4 successive yearly totals and dividing the result by 4. Thereafer, pairing the results and dividing them by 2, the most extreme fluctuations of the base data are smoothed.
The authors state“…a moving average has no effect on the period or length of the series being averaged. But it does have an effect upon the amplitude of the waves ”(p. 251)
Moving averages eliminate the appearent general randomness of data and show the basic rhythms of the process under study.