The SSM Procedure (Experimental)

controls the additive outlier search. The AOALPHA= option specifies the significance level for reporting the outliers. The default is 0.05. The MAXNUM= option limits the number of outliers to search. The default is MAXNUM=5. The MAXPCT= option is similar to the MAXNUM= option. In the MAXPCT= option you can limit the number of outliers to search for according to a percentage of the series length. The default is MAXPCT=1. When you specify both of these options, the minimum of the two search numbers is used.

specifies the significance level of the forecast confidence intervals. For example, ALPHA=0.05, which is the default, results in a 95% confidence interval.

causes the computation of the maximal state shock chi-square statistic at each distinct time point in the input data set. These statistics are output to the data set that is specified in the OUT= option. A time series plot of these statistics is produced if the PLOTS=MAXSHOCK option is specified in the PROC SSM statement. These statistics are useful for detecting structural breaks in the observation process. This option can be computationally expensive for a model with large state size.

specifies an output data set for the forecasts. The output data set contains the ID variable (if specified), the response variables, the one-step-ahead and out-of-sample response variable forecasts, the forecast confidence intervals, the smoothed values of the response series, and the one-step-ahead and smoothed estimates of the model components—including expressions that are defined by using the EVAL statement. See the section OUT= Data Set for more information.

causes the inclusion of the variables (variables in the program data vector) that are defined by using the programming statements in the SSM procedure in the OUT= data set. The parameters defined by the PARMS statement are also included. The output data set contains the values of these variables evaluated for all the rows in the input data set that is specified in the DATA= option. The parameters in the PARMS statement contain their estimated values.

prints the prediction error sum of squares (PRESS). The PRESS table reports the prediction error sum of squares, the number of summands used in the sum of squares, and the sum of squares of the weighted prediction errors (prediction errors normalized by their standard errors). See the section Smoothing Phase for more information about the precise meaning of the term prediction error.