The SGPANEL Procedure

PANELBY Statement

Specifies one or more classification variables for the panel, the layout type, and other options for the panel.

Syntax

PANELBY variable(s) </ option(s)>;

option(s) can be one or more of the following:

	BORDER \| NOBORDER
	COLHEADERPOS= TOP \| BOTTOM \| BOTH
	COLUMNS= n
	LAYOUT= LATTICE \| PANEL \| ROWLATTICE \| COLUMNLATTICE
	MISSING
	NOVARNAME
	ONEPANEL
	ROWHEADERPOS= RIGHT \| LEFT \| BOTH
	ROWS= n
	SPACING= n
	SPARSE
	START= TOPLEFT \| BOTTOMLEFT
	UNISCALE= ROW \| ALL

Required Arguments

variable(s): specifies one or more classification variables for the panel.

Options

BORDER | NOBORDER

specifies whether borders are displayed around each cell in the panel. BORDER adds the borders. NOBORDER removes the borders. Depending on the current ODS style, the borders might be present by default.

Style element:	The default status of the cell borders is determined by the FrameBorder attribute of the GraphWalls element in the current style.
Restriction:	This option is available with SAS 9.2 Phase 2 and later.

COLHEADERPOS= TOP | BOTTOM | BOTH

specifies the location of the column headings in the panel. Specify one of the following values:

TOP: places column headings at the top of each column.
BOTTOM: places column headings at the bottom of each column.
BOTH: places column headings at the top and bottom of each column.

Default:	TOP
Interaction:	This option has no effect if the panel uses the PANEL layout.
Restriction:	This option is available with SAS 9.2 Phase 2 and later.

COLUMNS= n

specifies the number of columns in the panel. By default, the number of columns is determined automatically based on the number of classifier values and the layout type.

Tip:	The SGPANEL procedure automatically splits the panel into multiple graphs (pages) as needed when your panel contains a large number of cells. You can control the number of cells in each graph by using the COLUMNS= and the ROWS= options.

LAYOUT= LATTICE | PANEL | COLUMNLATTICE | ROWLATTICE

specifies the type of layout that is used for the panel.

Select one of the following values:

LATTICE

when you specify two classification variables, arranges the cells so that the values of the first variable are columns and the values of the second variable are rows. You can use LATTICE only when you specify exactly two classification variables.

PANEL

arranges the cells in rows and columns. The headings for each cell are placed at the top of the cell.

COLUMNLATTICE

arranges the cells in a single row. You can use the COLUMNLATTICE layout only with a single classification variable.

Restriction:

This value is available with SAS 9.2 Phase 2 and later.

ROWLATTICE

arranges the cells in a single column. You can use the ROWLATTICE layout only with a single classification variable.

Restriction:

This value is available with SAS 9.2 Phase 2 and later.

Default:

PANEL

MISSING

processes missing values as a valid classification value and creates cells for it. By default, missing values are not processed as a classification value.

NOVARNAME

removes the variable names from the cell headings of a panel layout, or from the row and column headings of a lattice layout. For example, a row heading might "NorthEast" instead of "Region=NorthEast" when you specify the NOVARNAME option.

ONEPANEL

places the entire panel in a single output image. If you do not specify this option, then the panel is automatically split into multiple images as appropriate.

Note: This option is recommended only for panels with a small number of cells. If your panel is too large for the output image, then a blank image is created. [cautionend]

Interaction:	When you use ONEPANEL with the PANEL layout, only one of the ROWS= and COLUMNS= options can be used. If you specify both options, then the value for COLUMNS= is used. When you use ONEPANEL with the LATTICE layout, the ROWS= and COLUMNS= options have no effect.
Restriction:	This option is available with SAS 9.2 Phase 2 and later.

ROWHEADERPOS= LEFT| RIGHT | BOTH

specifies the location of the row headings in the panel. Specify one of the following values:

LEFT: places row headings at the left side of each row.
RIGHT: places row headings at the right side of each row.
BOTH: places row headings at the left side and the right side of each row.

Default:	TOP
Interaction:	This option has no effect if the panel uses the PANEL layout.
Restriction:	This option is available with SAS 9.2 Phase 2 and later.

ROWS= n

specifies the number of rows in the panel. By default, the number of rows is determined automatically based on the number of classifier values and the layout type.

Tip:	The SGPANEL procedure automatically splits the panel into multiple graphs (pages) as needed when your panel contains a large number of cells. You can control the number of cells in each graph by using the COLUMNS= and the ROWS= options.

SPACING= n

specifies the number of pixels between the rows and columns in the panel.

Default:

SPARSE

enables the SGPANEL procedure to create empty cells for crossings of the classification variables that are not present in the input data set. By default, empty cells are not created for the panel layout.

Interaction:

This option has no effect if you specify LAYOUT=LATTICE.

START= TOPLEFT | BOTTOMLEFT

specifies whether the first cell in the panel is placed at the upper left corner or the lower left corner. Specify one of the following values:

TOPLEFT

places the cell for the first data crossing in the upper left corner. Cells are placed from left to right, starting in the top row. Each additional row is placed below the previous row.

The following figure shows the placement of nine cells in a panel where START= TOPLEFT:

[Placement of cells with START=TOPLEFT]

BOTTOMLEFT

places the cell for the first data crossing in the lower left corner. Cells are placed from left to right, starting in the bottom row. Each additional row is placed above the previous row.

The following figure shows the placement of nine cells in a panel where START=BOTTOMLEFT:

[Placement of cells with START=BOTTOMLEFT]

Default:	TOPLEFT
Restriction:	This option is available with SAS 9.2 Phase 2 and later.

UNISCALE= COLUMN | ROW | ALL

scales the shared axes in the panel to be identical. Specify one of the following values:

COLUMN	scales all of the column axes in the panel to be identical.
ROW	scales all of the row axes in the panel to be identical.
ALL	scales all of the column axes to be identical, and also scales all of the row axes to be identical.

Default:

ALL

Top of Page