SMOOTH
The SMOOTH command provides options for smoothing time series by the methods of moving averages, exponential smoothing and seasonal adjustment.
In general, the format is:
SMOOTH var / options
The available options are:
| OPTION |
DESCRIPTION |
| additive |
Uses an additive model instead of the multiplicative model. If this option is used the ARITH option is automatically set. |
| arith |
Used to compute an arithmetic average rather than a geometric average for the calculation of seasonal factors. |
| beg= |
Specifies the BEGinning observation to be used in estimation. This option overrides the SAMPLE command and defaults to the sample range in effect. |
| central |
By default, centered moving averages are calculated. Use NOCENTRAL for fully lagged moving averages. |
| emave= |
Stores the exponential moving average in the variable specified. |
| end= |
Specifies the ENDing observation to be used in estimation. This option overrides the SAMPLE command and defaults to the sample range in effect. |
| hpfilter |
Implements the Hodrick-Prescott [1997] filter. The smoothing parameter is specified with the LAMBDA= option. |
| lambda= |
Specifies the smoothing parameter for the Hodrick-Prescott filter. |
| mave= |
Stores the moving average in the variable specified. |
| nma= |
Specifies the number of periods (p) to use in calculating moving averages. For seasonal adjustment NMA= is usually the same number as NSPAN=. |
| nocentral |
By default, centered moving averages are calculated. Use NOCENTRAL for fully lagged moving averages. |
| nspan= |
Specifies the number of seasonal periods S. Use 4 for quarterly data and 12 for monthly data. |
| samave= |
Stores the seasonally adjusted time series in the variable specified. |
| sfac= |
Stores the seasonal adjustment factors in the variable specified. |
| weight= |
Specifies the weight W in the range 0 < W < 1 in exponential smoothing to compute exponential moving averages. The default is 2/(1+p) where p is the value specified with the NMA= option. |
|
