The Moving Average command projects values in the forecast period, based on the average value of the variable over a specific number of preceding periods. Moving average is considered to be an effective tool to remove noise from time series and so reveal trends.

How To

Run: Statistics→Time Series → Moving Average...

Select a variable containing a time series.

Select a moving average technique – simple, centered, weighted or Spencer's (v6 and newer).

Results

Table of the predicted values and the errors is produced. Moving average values are placed at the period in which they are calculated. #N/A value is shown when there are insufficient historical values to project a forecast or calculate a standard error.

Methods

A simple moving average is the unweighted mean of the consecutive data points. Default length for the moving average is 3.

References

[ENW] Enders, Walter (2004). "Stationary Time-Series Models". Applied Econometric Time Series (Second ed.). New York: Wiley.