1. Home
  2. Moving Average Options

Moving Average Options

This dialog allows the options for the Moving Average menu to be set.

Display

If Holt-Winters parameters is set, then the estimated parameters found by optimization for the Holt-Winters filter (alpha, beta and gamma) will be displayed in the output.

Graph title

Lets you specify a title to use for all the graphs. If this is left blank, default titles are created. If you want to suppress the title just enter a space in the title field.

Graphics

This lets you control which graphs are plotted:

Moving averages A graph with the observed and fitted moving averages will be plotted using a line plot.
Individual components A graph with the trend, level and seasonal components will be plotted for a Holt-Winters filter or where there is a seasonal adjustment fitted.
Predictions for Holt-Winters filter A graph giving the predicted values at the end of the series for a Holt-Winters filter will be plotted with 95% confidence limits. The number of points that will be predicted are given by the Number of predictions for Holt-Winters filter field.

Number of predictions for Holt-Winters filter

This provides a space to supply the number of future points at the end of the series that will be predicted for a Holt-Winters filter.

Holt-Winters filter

This allows the specification of parameters of the Holt-Winters filter. The smoothing parameters Alpha, Beta and Gamma control the balance between past and current contributions to the estimates of the level, trend and seasonal components respectively. They must take a value between 0 and 1, and the closer they are to zero, the less the contribution of the current value to the estimate, giving greater smoothing of the series. If any of them are unset, then optimal smoothing parameters for the given dataset will be found. If Multiplicative seasonal model is set, then a multiplicative model for seasonal effects is used (i.e. estimate = (level + trend)*season), otherwise an additive model is used (i.e. estimate = level + trend + season).

Multiplicative seasonal model

Multiplicative seasonal model: When selected the seasonal component is multiplicative in the Holt-Winters method, otherwise it is additive model.

Action Icons

Clear Clear all fields and list boxes.
Help Open the Help topic for this dialog.

See also

Updated on March 22, 2019

Was this article helpful?