The TIMESERIES Procedure

specifies whether the series should be adjusted by its mean prior to performing the singular spectrum analysis. You can specify the following values:

By default, ADJUSTMEAN=NO.

specifies the lists that categorize window lags into groups. The window lags must be separated by spaces or commas. For example, GROUPS=(1,3) (2,4) specifies that the first and third window lags form the first group and the second and fourth window lags form the second group. If you do not specify the GROUPS= option, the window lags are divided into two groups based on the THRESHOLDPCT= value.

For example, the following SSA statement specifies three groups:

   ssa / groups=(1 3)(2 4 5)(6);

The first group contains the first and third principal components; the second group contains the second, fourth, and fifth principal components; and the third group contains the sixth principal component.

By default, the first group contains the principal components whose contributions to the series sum to greater than the THRESHOLDPCT= value of 90%, and the second group contains the remaining components.

specifies the window length to be used in the analysis. The window length represents the maximum lag to be used in the SSA autocovariance calculations, where number must be greater than 1. When the SEASONALITY= option is provided or implied by the INTERVAL= option in the ID statement, the default window length is the smaller of two times the length of the seasonal cycle and one-half the length of the time series. When no seasonality value is available, the default window length is the smaller of 12 and one half the length of the time series.

For example, the following SSA statement specifies a window length of 10:

   ssa / length=10;

If the specified number is greater than one-half the length of the accumulated time series, the window length is reduced and a warning message is printed to the log. If you do not specify the window length option and the INTERVAL=MONTH or SEASONALITY=12 options are specified, a window length of 24 is used.

specifies the number of time periods to be stored in the OUTSSA= data set when the TRANSPOSE=YES option is specified. If the TRANSPOSE option is not specified, the NPERIODS= option is ignored. The NPERIODS= option specifies the number of OUTSSA= data set variables to contain the groups.

If you do not specify this option, all the periods specified between the START= and END= options are stored of the ID statement. If at least one of the START= or END= options is not specified, the default magnitude is the seasonality specified by the SEASONALITY= option in the PROC TIMESERIES statement or implied by the INTERVAL= option in the ID statement. If only the START= option or both the START= and END= options are specified and the seasonality is zero, the default is NPERIODS=5. If only the END= option or neither the START= nor END= option is specified and the seasonality is zero, the default is NPERIODS=–5.

specifies a percentage to be used to divide the SSA components into two groups based on the cumulative percentage of their singular values. The percentage must be between 0 and 100, inclusive. By default, THRESHOLDPCT=90.

For example, the following SSA statement specifies 80%:

   ssa / THRESHOLDPCT=80;

The size of the second group must be at least 1, and it must be less than the window length. The percentage is adjusted to achieve this requirement.

For example, the following SSA statement specifies THRESHOLDPCT=0, which effectively sets the size of the second group to one less than the window length:

   ssa / THRESHOLDPCT = 0;

The following SSA statement specifies 100%, which implies that the size of the last group is one:

   ssa / THRESHOLDPCT= 100;

specifies which values are recorded as column names in the OUTSSA= data set.

By default, TRANSPOSE=NO.