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 ModelsModels with Dependent LagsCovariance 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
Syntax: SSM Procedure
The following statements are available in the SSM procedure:
-
PROC SSM <options>;
-
BY variables;
-
COMPONENT name = (variables) * state </ options>;
-
DEPLAG
name(response-variable) lag-term1 <lag-term2 …> ; -
EVAL name = expression </ options>;
-
ID variable <option>;
-
IRREGULAR name <options>;
-
MODEL response = variables </ options>;
-
OUTPUT <options>;
-
PARMS variables </ options> ;
-
STATE name(dim) <options>;
-
TREND name(type) <options>;
You can specify all statements except the BY, ID, and the OUTPUT statements multiple times. The PROC SSM statement and at least one MODEL statement are required. In addition to these statements, you can use most DATA step programming statements to define new variables that are needed for specifying different parts of the state space model.
Note: In the statement options described throughout this section, whenever you use a list to specify the elements of the system matrices, the list elements must all be of the same type: either all of them must be variables or all of them must be numbers. In addition, if the list contains more than one variable, then they cannot be of the array type. These are not serious restrictions. When the list contains mix of variables and numbers, you can redefine the numbers as constant variables. Similarly, you can reformulate a list that contains a mix of variables of array and non-array types as just one array by combining all its elements in a new array.