Procedure Syntax

Requirements: Exactly one ID statement is required.

Reminder: The procedure can include the BY statement.

PROC GREDUCE < option(s) >;

• ID id-variable(s) ;

PROC GREDUCE Statement

Identifies the input and output map data sets. Optionally specifies the reduction criteria.

Requirements: An input map data set is required.

Syntax

PROC GREDUCE < option(s) >;

option(s) can be one or more options from any or all of the following categories:

• data set options:

  DATA= input-map-data-set

  OUT= output-map-data-set

• level options:

  E1= min-distance

  E2= min-distance

  E3= min-distance

  E4= min-distance

  E5= min-distance

  N1= max-points

  N2= max-points

  N3= max-points

  N4= max-points

  N5= max-points

Options

DATA= input-map-data-set

• identifies the map data set that you want to process. By default, the procedure uses the most recently created SAS data set.

• See also: About the Input Map Data Set on page 1215 and SAS Data Sets on page 29.

E1= min-distance

E2= min-distance

E3= min-distance

E4= min-distance

E5= min-distance

• specify the minimum distance that a point must lie from a straight line segment to be included at density level 1, 2, 3, 4, or 5, respectively. That is, in a reduced curve of three points, the middle point is at least a distance that is min-distance from a straight line between the two outside points.

  Express min-distance values in the units for the coordinate system of the input map data set. For example, if the input map data set contains coordinates that are expressed in radians, express the min-distance values in radians.

  Specify the E n = values in decreasing order. For example, the E2= value should be less than the E1= value and so on.

N1= max-points

N2= max-points

N3= max-points

N4= max-points

N5= max-points

• specify that for density level 1, 2, 3, 4, or 5, the boundary of a unit area should contain no more than max-points points.

  Specify the N n = values in increasing order. For example, the N2= value should be greater than or equal to the N1= value and so on.

  By default, if you omit N n = and E n = , the GREDUCE procedure calculates values for the five N n = parameters using this formula:

  

  Here N _max is the maximum number of points in any unit area in the input map data set. However, the restriction that the number of points for any level cannot be less than the number of points in level 0 still applies.

OUT= output-data-set

• names the new map data set, which contains all of the observations and variables in the original map data set plus the new DENSITY variable. If the input map data set contains a variable named DENSITY, the GREDUCE procedure replaces the values of the variable in the output map data set. The original values of the DENSITY variable from the input map data set are not included in the output map data set.

  By default, the GREDUCE procedure names the new data set that uses the DATA n naming convention. That is, the procedure uses the name WORK.DATA n , where n is the next unused number in sequence. Thus, the first automatically named data set is DATA1, the second is DATA2, and so on.

ID Statement

Identifies the variable or variables that define the hierarchy of the current unit areas in the input map data set.

Requirements: At least one id-variable is required.

Featured in: Example 1 on page 1220.

Syntax

• ID id-variable(s) ;

Required Arguments

id-variable(s)

• specifies one or more variables in the input map data set that identify unit areas. Id-variable(s) can be either numeric or character.

  Each group of observations with a different ID variable value is evaluated as a separate unit area.