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 group number between 1 and 9999, inclusively.
The GROUP statement signifies the beginning of a group specification block and designates a group number for the group. All subsidiary group specification statements after a GROUP statement belong in that group until another MODEL or GROUP statement is used. The subsidiary group specification statements refer to one of the following four statements:
For example, consider the following statements:
proc calis;
var X1-X4;
group 1 / label='Women' data=women_data;
freq Z;
group 2 / label='Men' data=men_data;
partial P;
model 1 / group = 1-2;
factor N=1; /* One factor exploratory factor analysis */
run;
In the GROUP statements, two groups are defined. Group 1, labeled as 'Women', refers to the data set women_data. Group 2, labeled as 'Men', refers to the data set men_data. Both groups are fitted by an exploratory factor model defined in Model 1, as indicated in the GROUP= option of the MODEL
statement. While the frequency variable Z defined in the FREQ
statement is applicable only to Group 1, the partial variable P defined in the PARTIAL
statement is applicable only to Group 2. However, the VAR
statement, which appears before the definitions of both groups, applies globally to both Group 1 and Group 2. Therefore,
variables X1–X4 are the analysis variables in the two groups.
You can set group-specific options in each GROUP statement. All but one (that is, the LABEL= option) of these options are also available in the MODEL and PROC CALIS statements. If you set these group-specific options in the PROC CALIS statement, they will apply to all groups unless you respecify them in the GROUP statement. If you set these group-specific options in the MODEL statement, they will apply to all groups that are fitted by the associated model. In general, the group-specific options are transferred from the PROC CALIS statement to the MODEL statements (if present) and then to the fitted groups. In the transferring process, options are overwritten by the newer ones. If you want to apply some group-specific options to a particular group only, you should set those options in the GROUP statement corresponding to that group.
These options are available in the GROUP and PROC CALIS statements:
|
Option |
Description |
|---|---|
|
Specifies the input data set |
|
|
Specifies the data set that contains the weight matrix |
|
|
Specifies the data set for storing the statistical results |
|
|
Specifies the data set for storing the weight matrix |
|
|
Specifies the maximum number of iterations for estimating robust covariance and mean matrices |
|
|
Specifies the tuning parameter for robust methods |
See the section PROC CALIS Statement Options for more details about these options. If you specify these options in the PROC CALIS statement, they are transferred to all GROUP statements. They might be overwritten by the respecifications in the individual GROUP statements.
These options are available in the GROUP, MODEL , and PROC CALIS statements:
|
Option |
Description |
|---|---|
|
Specifies the |
|
|
Specifies the |
|
|
Computes the skewness and kurtosis without bias corrections |
|
|
Defines nobs by the number of error df |
|
|
Specifies that the INWGT= data set contains the inverse of the weight matrix |
|
|
Computes and displays kurtosis |
|
|
Specifies the maximum number of missing patterns to display |
|
|
Defines the number of observations (nobs) |
|
|
Suppresses the display of missing pattern analysis |
|
|
Displays analyzed and estimated moment matrix |
|
|
Specifies ODS Graphics selection |
|
|
Displays the weight matrix |
|
|
Defines nobs by the number of regression df |
|
|
Computes the default residuals |
|
|
Specifies the type of residuals |
|
|
Specifies the ridge factor for covariance matrix |
|
|
Prints univariate statistics |
|
|
Specifies the data proportion threshold for displaying the missing patterns |
|
|
Specifies variance divisor |
|
|
Specifies the penalty weight to fit correlations |
|
|
Specifies the ridge factor for the weight matrix |
If you specify these options in the PROC CALIS statement, they are transferred to all MODEL statements. These options are overwritten by the respecifications in the individual MODEL statements. After these options are resolved in a given MODEL statement, they are transferred further to the GROUP statements of which the associated groups are fitted by the model. Again, these options might be overwritten by the respecifications in the individual GROUP statements.