Fit Model Script-Only Messages

The following messages for a Model Dialog object are available only in JSL:

model_dialog_object << Cauchy Fit( 0|1 ); //Response Screening

model_dialog_object << Choose High Target( 0|1 ); //Nominal Logistic

model_dialog_object << Generalized Distribution( string ); //Generalized Regression

model_dialog_object << GLM Distribution( string ); //Generalized Linear Model

model_dialog_object << Max Iterations( number ); //Partial Least Squares

model_dialog_object << Power Link Parameter( number ); //Generalized Linear Model

model_dialog_object << Robust Fit( 0|1 ); //Response Screening

model_dialog_object << Suppress Warnings for Missing Effects;

The following sections describe scripting-only messages for MANOVA, Generalized Linear Models, Generalized Regression, Nominal and Ordinal Logistic, Standard Least Squares, and Stepwise.

MANOVA

The following message for a Fit MANOVA object is available only in JSL:

fit_model_object << ( Response[1] << (Effect[1] << ... ) );

The third message can be one of the following:

Test Details // show or hide the test details for an individual effect

Centroid Plot // show or hide the centroid plot for an individual effect

Save Canonical Scores // save the canonical scores for an individual effect

Contrast /* run a customized F test contrasting different levels for

			an effect in the model*/

The following example saves canonical scores for the Response Function, which in this data, is Sum:

dt = Open( "$SAMPLE_DATA/Dogs.jmp" );

obj = dt << Fit Model(

	Y( :LogHist0, :LogHist1, :LogHist3, :LogHist5 ),

	Effects( :drug, :dep1, :drug * :dep1 ),

	Personality( "Manova" ),

	Run( Response Function( Sum ) )

);

obj << (Response[1] << (Effect[1] << Save Canonical Scores) );

Generalized Linear Models

The following messages for a Fit Generalized Linear Model object are available only in JSL:

fit_model_object << GLM Distribution( distribution );

fit_model_object << Parametric Formula;

fit_model_object << Power Link Parameter( number );

Generalized Regression

The following messages for a Fit Generalized Regression object are available only in JSL:

fit_model_object << Fit( ... );

fit_model_object << Get X Matrix;

The following messages for a Fit Generalized Regression object are available only in JSL and must be sent to an individual model fit. The model fits are numbered sequentially in the order they appear in the report.

fit_model_object << (Fit[n] << Get Prediction Formula);

fit_model_object << (Fit[n] << Set Solution ID( number ));

fit_model_object << (Fit[n] << Step Backward);

fit_model_object << (Fit[n] << Step Forward);

Nominal and Ordinal Logistic

The following messages for Fit Nominal Logistic and Fit Ordinal Logistic objects are available only in JSL:

fit_model_object << Get Confusion Matrix Test;

fit_model_object << Get Confusion Matrix Training;

fit_model_object << Get Confusion Matrix Validation;

fit_model_object << Get Confusion Rates Test;

fit_model_object << Get Confusion Rates Training;

fit_model_object << Get Confusion Rates Validation;

fit_model_object << Get MM SAS Data Step;

fit_model_object << Get SAS Data Step;

fit_model_object << Get Measures;

fit_model_object << Get Probability Formulas;

The following message for a Fit Nominal Logistic object is available only in JSL:

fit_model_object << Positive Level( level );

Standard Least Squares

The following messages for a Fit Least Squares object are available only in JSL:

fit_model_object << Get Conditional Formula;

fit_model_object << Get Effect Names;

fit_model_object << Get Effect PValues;

fit_model_object << Get Estimates;

fit_model_object << Get Indiv Confid Limit Formula;

fit_model_object << Get Mean Confid Limit Formula;

fit_model_object << Get Measures;

fit_model_object << Get MM SAS DATA Step;

fit_model_object << Get Parameter Names;

fit_model_object << Get Parameterized Formula;

fit_model_object << Get Prediction Formula;

fit_model_object << Get Random Effect Names;

fit_model_object << Get Std Errors;

fit_model_object << Get Standard Error Formula;

fit_model_object << Get Variance Components;

fit_model_object << Get X Matrix;

fit_model_object << Get XPX Inverse;

fit_model_object << Get Y Matrix;

fit_model_object << Profiler( Extrapolation Control Criterion( string ) );

fit_model_object << Show VIF;

Stepwise

The following messages for a Fit Stepwise object are available only in JSL:

fit_stepwise_object << Backward Step;

fit_stepwise_object << Enter( column );

fit_stepwise_object << Finish; // preferred in scripts over << Go

fit_stepwise_object << Forward Step;

fit_stepwise_object << Get Measures;

fit_stepwise_object << Lock( column );

fit_stepwise_object << Prob to Enter( number );

fit_stepwise_object << Prob to Leave( number );

fit_stepwise_object << Remove( column );

fit_stepwise_object << Unlock( column );

需要更多信息？有问题？从 JMP 用户社区得到解答 (community.jmp.com).