• OPTION is the PROC CLUSTER option,
Omits points with low estimated probability densities from the analysis. Valid values for n are 0 < n < 100. If n < 1, then n is the proportion of observations omitted. If n > 1, then n is interpreted as a percentage. A specification of TRIM=10, which trims 10% of the points, is a reasonable value for many data sets. K=n Specifies the number of neighbors to use for kth-nearest-neighbor density estimation. The number of neighbors (n) must be at least two but less than the number of observations. R=n Specifies the radius of the sphere of support (n) for uniform-kernel density estimation. The value of n must be greater than zero. Requests the hybrid clustering method in which density estimates are computed from a preliminary cluster analysis using the K-Means clustering method. Specifies the beta parameter (n) for the flexible beta method. The value of n should be less than 1, usually between 0 and -1. By default, BETA= -0.25.
Type a specific PROC CLUSTER option in the Additional PROC CLUSTER Options field.
For example, to specify a trim value of 5%, type TRIM=5 in the text field, as shown below.Refer to the SAS PROC CLUSTER documentation for more information.