The PROBIT Procedure

Overview

You can specify interaction terms among any of the specified explanatory variables in the MODEL statement in the same way as in the GLM procedure. Because of the more complicated models that you can specify, only a single MODEL statement is now allowed for each PROC PROBIT statement. If more than one MODEL statement is specified, only the last one is used.

You can now graphically display results of model parameter estimation. For binary data analysis, you can construct plots of predicted probabilities, inverse predicted probabilities, and cumulative probabilities. For ordinal multinomial models, you can display plots of predicted probabilitites, linear predictors, and cumulative probabilities. In addition, you can control graphical features such as plot layout, colors, plotting symbols, line styles, and the text fonts used in the plots.

You can use the new XDATA= data set to provide the values for effects other than the single continuous dosage effect when the predicted dosage and their fiducial limits are computed. These specified values can also be used for generating plots.

You can use the new INEST= data set to provide the initial values for parameters in the model. The structure of this data set is similar to that of the OUTEST= data set. This data set will overwrite the original INITIAL= option in the MODEL statement.

The OUTEST= data set is extended to both binomial and multinomial models. It also generates the OUTEST data set whether there is a CLASS statement or not. The OUTEST= data set does not include a fixed threshold.

You can use the new AGGREGATE= option in the MODEL statement to define the subpopulations for calculating the Pearson chi-square statistic and the deviance.

Chapter Contents
Previous
Next
Top