The CALIS Procedure
-
Overview
-
Getting Started
-
SyntaxClasses of Statements in PROC CALISSingle-Group Analysis SyntaxMultiple-Group Multiple-Model Analysis SyntaxPROC CALIS StatementBOUNDS StatementBY StatementCOSAN StatementCOV StatementDETERM StatementEFFPART StatementFACTOR StatementFITINDEX StatementFREQ StatementGROUP StatementLINCON StatementLINEQS StatementLISMOD StatementLMTESTS StatementMATRIX StatementMEAN StatementMODEL StatementMSTRUCT StatementNLINCON StatementNLOPTIONS StatementOUTFILES StatementPARAMETERS StatementPARTIAL StatementPATH StatementPATHDIAGRAM StatementPCOV StatementPVAR StatementRAM StatementREFMODEL StatementRENAMEPARM StatementSAS Programming StatementsSIMTESTS StatementSTD StatementSTRUCTEQ StatementTESTFUNC StatementVAR StatementVARIANCE StatementVARNAMES StatementWEIGHT Statement
-
DetailsInput Data SetsOutput Data SetsDefault Analysis Type and Default ParameterizationThe COSAN ModelThe FACTOR ModelThe LINEQS ModelThe LISMOD Model and SubmodelsThe MSTRUCT ModelThe PATH ModelThe RAM ModelNaming Variables and ParametersSetting Constraints on ParametersAutomatic Variable SelectionPath Diagrams: Layout Algorithms, Default Settings, and CustomizationEstimation CriteriaRelationships among Estimation CriteriaGradient, Hessian, Information Matrix, and Approximate Standard ErrorsCounting the Degrees of FreedomAssessment of FitCase-Level Residuals, Outliers, Leverage Observations, and Residual DiagnosticsTotal, Direct, and Indirect EffectsStandardized SolutionsModification IndicesMissing Values and the Analysis of Missing PatternsMeasures of Multivariate KurtosisInitial EstimatesUse of Optimization TechniquesComputational ProblemsDisplayed OutputODS Table NamesODS Graphics
-
ExamplesEstimating Covariances and CorrelationsEstimating Covariances and Means SimultaneouslyTesting Uncorrelatedness of VariablesTesting Covariance PatternsTesting Some Standard Covariance Pattern HypothesesLinear Regression ModelMultivariate Regression ModelsMeasurement Error ModelsTesting Specific Measurement Error ModelsMeasurement Error Models with Multiple PredictorsMeasurement Error Models Specified As Linear EquationsConfirmatory Factor ModelsConfirmatory Factor Models: Some VariationsResidual Diagnostics and Robust EstimationThe Full Information Maximum Likelihood MethodComparing the ML and FIML EstimationPath Analysis: Stability of AlienationSimultaneous Equations with Mean Structures and Reciprocal PathsFitting Direct Covariance StructuresConfirmatory Factor Analysis: Cognitive AbilitiesTesting Equality of Two Covariance Matrices Using a Multiple-Group AnalysisTesting Equality of Covariance and Mean Matrices between Independent GroupsIllustrating Various General Modeling LanguagesTesting Competing Path Models for the Career Aspiration DataFitting a Latent Growth Curve ModelHigher-Order and Hierarchical Factor ModelsLinear Relations among Factor LoadingsMultiple-Group Model for Purchasing BehaviorFitting the RAM and EQS Models by the COSAN Modeling LanguageSecond-Order Confirmatory Factor AnalysisLinear Relations among Factor Loadings: COSAN Model SpecificationOrdinal Relations among Factor LoadingsLongitudinal Factor Analysis
- References
where i is an assigned model number between 1 and 9999, inclusively.
A MODEL statement signifies the beginning of a model specification block and designates a model number for the model. All main and subsidiary model specification statements after a MODEL statement belong in that model until another MODEL or GROUP statement is encountered.
The MODEL statement itself does not serve the purpose of model specification, which is actually done by the main and subsidiary model specification statements that follow it. The MODEL statement serves as a "place-holder" of specification of a single model. It also makes the reference to a model easier with an assigned model number. For example, consider the following statements:
proc calis;
group 1 / data=women_data;
group 2 / data=men_data;
model 1 / group=1 label='Women Model';
{model 1 specification here}
model 2 / group=2 label='Men Model';
{model 2 specification here}
run;
This example illustrates a two-group analysis with two models. One is model 1 labeled as 'Women Model' in a MODEL statement. Another is model 2 labeled as 'Men Model' in another MODEL statement. The two groups, group 1 and group 2, as defined in two separate GROUP
statements, are fitted by model 1 and model 2, respectively, as indicated by the GROUP= option of the MODEL statements. Within the scope of model 1, you provide model specification statements by using the main
and subsidiary
model specification statements. Usually, one of the following main model specification statements is used: FACTOR
, LINEQS
, LISMOD
, MSTRUCT
, PATH
, RAM
, or REFMODEL
. Similarly, you provide another set of model specification statements within the scope of model 2.
Hence, for an analysis with a single group, the use of the MODEL statement is not necessary because the model that fits the group is unambiguous.
You can set model-specific options in each MODEL statement. All but two of these options are also available in the PROC CALIS statement. If you set these options in the PROC CALIS statement, they apply to all models, unless you respecify them in the local MODEL statements. If you want to apply some options only to a particular model, specify these options in the MODEL statement that corresponds to that model.
You can also set group-specific options in the MODEL statement. These group options apply to the groups that are specified in GROUP= option of the MODEL statement. See the section Options Available in the GROUP and PROC CALIS Statements for a detailed descriptions of these group options.
The following options are available in the MODEL and PROC CALIS statements. If you specify these options in the PROC CALIS statement, they are transferred to all MODEL statements. These options might be overwritten by the respecifications in the local MODEL statements.
|
Option |
Description |
|---|---|
|
Emphasizes the diagonal entries |
|
|
Displays total, direct, and indirect effects |
|
|
Displays the extended path estimates that include variances and covariances |
|
|
Specifies the data set that contains the initial values and constraints |
|
|
Specifies the data set that contains the model specifications |
|
|
Analyzes the mean structures |
|
|
Computes modification indices |
|
|
Deactivates the inherited MEANSTR option |
|
|
Suppresses modification indices |
|
|
Displays model specifications and results according to the input order |
|
|
Suppresses the printing of parameter names in results |
|
|
Suppresses the standardized output |
|
|
Suppresses standard error computations |
|
|
Orders the model output displays according to the parameter types within each model |
|
|
Specifies the data set that outputs the estimates and their covariance matrix |
|
|
Specifies the output data set for storing the model specification and results |
|
|
Displays parameter names in model specifications and results |
|
|
Computes the determination coefficients |
|
|
Prints parameter estimates |
|
|
Prints initial pattern and values |
|
|
Computes the latent variable covariances and scoring coefficients |
|
|
Displays results in matrix forms |
|
|
Instructs generated default parameters be read in the INMODEL= data set |
|
|
Computes the standard errors |
Some options in the GROUP statement can also be specified in the MODEL statements. Group options that are specified the MODEL statements are transferred to the GROUP statements that define the groups that are fitted by the associated models in the MODEL statements. This is a little more convenient than setting the common group options individually in the GROUP statements for all fitted groups by a model. See the section Options Available in GROUP, MODEL, and PROC CALIS Statements for a reference of these options.