The GCONTOUR Procedure

Concepts

[GCONTOUR Procedure Terms]

The GCONTOUR procedure requires three numeric variables to produce a plot. The input data set forms a rectangular grid from the values of x and y. The z variable is plotted on the grid as the third dimension. Only one value of z is required for each (x,y) grid location. If multiple observations have the same z value for any (x,y) combination, only the last observation is plotted.

Data Ranges

PROC GCONTOUR produces a rectangular grid with axes scaled to include the minimum data values and maximum data values of x and y. Each axis is labeled with the variable name or label. The contour lines represent the levels of magnitude by grouping the common values of the z variable. The level of each contour line is displayed in the legend. The legend label is the z variable's name or label.

Missing Values

PROC GCONTOUR requires data values for at least fifty percent of the z variable, for each unique combination of (x,y). The INCOMPLETE option can be used to override this requirement. The G3GRID procedure can also be used to create data for missing values. (See The G3GRID Procedure).

Interpolating Data

The G3GRID procedure enables you to produce a data set with nonmissing values for z for every unique (x,y) combination. The output data set from the G3GRID procedure can be used as the input data set for the GCONTOUR procedure. The G3GRID procedure also enables you to smooth data for use with GCONTOUR. For details see The G3GRID Procedure. For an interpolation example see Using the Default Interpolation Method .

Top of Page