Introduction to the SGPANEL Procedure

The SGPANEL procedure creates a panel for the values of one or more classification variables. Each graph cell in the panel can contain either a single plot or multiple overlaid plots. The procedure syntax supports the following features:

• two types of panel layouts: PANEL and LATTICE

• basic plots: scatter plots, series plots, band plots, and needle plots

• fit and confidence plots: loess curves, regression curves, and penalized B-spline curves

• distribution plots: histograms, box plots, and density curves

• categorization plots: bar charts, dot plots, and bar-line charts

• legends and reference lines

The SGPANEL procedure can create two different layouts, depending on the value of the LAYOUT= option. You can specify either the PANEL layout or the LATTICE layout.

The following examples show the types of layouts that you can create with the SGPANEL procedure.

This is an example of the default PANEL layout. In the PANEL layout, each graph cell represents a specific crossing of values for one or more classification variables. A label above each cell identifies the crossing of values that is represented in the cell. By default, cells are created only for crossings that are represented in the data set.

This is an example of the LATTICE layout. In the LATTICE layout, the graph cells are arranged in rows and columns by using the values of two classification variables. Labels above each column and to the right of each row identify the classification value that is represented by that row or column. A cell is created for each crossing of classification values.

For more information about the SGPANEL procedure and the procedure syntax, see The SGPANEL Procedure.

Copyright © 2007 by SAS Institute Inc., Cary, NC, USA. All rights reserved.