Shared Concepts and Topics: Syntax :: SAS/STAT(R) 9.3 User's Guide

Syntax: EFFECTPLOT Statement

EFFECTPLOT <plot-type<(plot-definition-options)>></ options> ;

The available plot-types and their plot-definition-options are described in Table 19.11. Table 19.13 lists the options that can be specified after a slash (/) for any plot-type, and Table 19.14 lists additional options that enhance specific plot-types. Full descriptions of the plot-definition-options and the other options are provided in the section Dictionary of Options.

Table 19.11 *Plot-Types* and *Plot-Definition-Options*
Plot-Type and Description	Plot-Definition-Options
BOX
Displays a box plot of continuous response data at each level of a CLASS effect, with predicted values superimposed and connected by a line. This is an alternative to the INTERACTION plot-type.	PLOTBY= variable or CLASS effect X= CLASS variable or effect
CONTOUR
Displays a contour plot of predicted values against two continuous covariates.	PLOTBY= variable or CLASS effect X= continuous variable Y= continuous variable
FIT
Displays a curve of predicted values versus a continuous variable.	PLOTBY= variable or CLASS effect X= continuous variable
INTERACTION
Displays a plot of predicted values (possibly with error bars) versus the levels of a CLASS effect. The predicted values are connected with lines and can be grouped by the levels of another CLASS effect.	PLOTBY= variable or CLASS effect SLICEBY= variable or CLASS effect X= CLASS variable or effect
SLICEFIT
Displays a curve of predicted values versus a continuous variable grouped by the levels of a CLASS effect.	PLOTBY= variable or CLASS effect SLICEBY= variable or CLASS effect X= continuous variable

By default, a single plot is produced based on the type of response variable and the number of continuous and classification covariates in the model as shown in Table 19.12. If you have a polytomous response model, then the response variable is treated as the grouping classification variable in this table. If your model does not fit into Table 19.12, then a default plot is not produced; however, specifying the plot-type argument displays a plot with the extra continuous covariates fixed at their mean values and the extra classification covariates fixed at their reference levels.

Table 19.12 Default *Plot-Types*
Number of Covariates		Type of Response Variable
Classification	Continuous	Continuous or Binary	Polytomous
1	0	INTERACTION	INTERACTION with groups
2	0	INTERACTION with groups	None
0	1	FIT	SLICEFIT
0	2	CONTOUR	None
1	1	SLICEFIT	None

Table 19.13 and Table 19.14 list the options that can be specified after a slash (/) to enhance the effect plots.

Table 19.13 Available *Options* for All *Plot-Types*
AT<args>	ATLEN=	ATORDER=	ILINK	INDIVIDUAL $\text{[math]}$
LINK	MOFF	NCOLS= $\text{[math]}$	NOOBS $\text{[math]}$	NROWS= $\text{[math]}$
OBS<(options)>	PLOTBYLEN=	PREDLABEL=	UNPACK
$\text{[math]}$ Not available for the BOX plot-type

Note: If your model contains an offset variable and the MOFF option is not specified or not valid, then the predicted values are computed only at the observations. In this case, the FIT and SLICEFIT plot-types display scatter plots of the predicted values, the CONTOUR plot-type displays the residuals against two continuous covariates but with no fitted surface, the INTERACTION plot-type does not connect the predicted values with lines, and the BOX plot-type is unchanged.

Table 19.14 Additional *Options* for Each *Plot-Type*
Plot-Type	Options
BOX	CLUSTER	YRANGE=
CONTOUR	EXTEND=	GRIDSIZE=
FIT	ALPHA=	EXTEND=	GRIDSIZE=	NOCLI
	NOCLM	NOLIMITS	SMOOTH	YRANGE=
INTERACTION	ALPHA=	CLI	CLM	LIMITS
	POLYBAR	YRANGE=
SLICEFIT	ALPHA=	CLI	CLM	EXTEND=
	GRIDSIZE=	LIMITS	YRANGE=

Dictionary of Options