The SSM Procedure
-
Overview
- Getting Started
-
Syntax
-
DetailsState Space Model and NotationTypes of Sequence DataOverview of Model Specification SyntaxFiltering, Smoothing, Likelihood, and Structural Break DetectionEstimation of User-Specified Linear Combination of State ElementsContrasting PROC SSM with Other SAS Procedures Predefined Trend ModelsPredefined Structural ModelsCovariance ParameterizationMissing ValuesComputational IssuesDisplayed OutputODS Table NamesODS Graph NamesOUT= Data Set
-
ExamplesBivariate Basic Structural Model Panel Data: Random-Effects and Autoregressive ModelsBackcasting, Forecasting, and InterpolationLongitudinal Data: Smoothing of Repeated MeasuresA User-Defined Trend ModelModel with Multiple ARIMA ComponentsDynamic Factor ModelingDiagnostic Plots and Structural Break AnalysisLongitudinal Data: Variable Bandwidth SmoothingA Transfer Function Model for the Gas Furnace DataPanel Data: Dynamic Panel Model for the Cigar DataMultivariate Modeling: Long-Term Temperature TrendsBivariate Model: Sales of Mink and Muskrat FursFactor Model: Now-Casting the US EconomyLongitudinal Data: Lung Function Analysis
- References
You can use the OUT= option in the OUTPUT statement to store the series and component forecasts that are produced by PROC SSM. Which columns are included in the data set depends on the model specification. The model can have one or more response variables, a variety of components that appear in the MODEL statement, and components specified by the EVAL statement. The OUT= data set contains the one-step-ahead and full-sample estimates of the response variables, and all these components.
The following list describes the columns of the data set:
-
the BY variables
-
the ID variable, if specified by the ID statement
-
Obs, a variable that contains the observation number
-
the response series (more than one in the multivariate case)
-
the following columns associated with the response series (the wildcard * is substituted by the name of one of the response variables):
-
FORECAST_* contains the one-step-ahead predicted values. and the multistep forecasts of the response series.
-
RESIDUAL_* contains the difference between the actual and forecast values.
-
StdErr_* contains the standard error of prediction.
-
Lower_* and Upper_* contain the lower and upper forecast confidence limits.
-
Smoothed_* contains the smoothed values of the response variable.
-
StdErr_Smoothed_* contains standard errors of the smoothed values of the response variable.
-
AO_* contains the additive outlier estimate.
-
StdErr_AO_* contains the standard error of the additive outlier estimate.
-
-
the following columns associated with the components (the wildcard * is substituted by the name of one of the components):
-
FORECAST_* contains the one-step-ahead predicted values and the multistep forecasts of the component.
-
StdErr_* contains the standard error of prediction.
-
Smoothed_* contains the smoothed values of the component.
-
StdErr_Smoothed_* contains standard errors of the smoothed values of the component.
-
Smoothed_Lower_* and Smoothed_Upper_* contain the lower and upper confidence limits of the smoothed component.
-
-
the maximal state shock chi-square statistics at distinct time points (this column is present only if the MAXSHOCK option is used in the OUTPUT statement)
Confidence limits are not produced for the smoothed series values or for the component forecasts; they are produced for the smoothed components.