topic Nearest neighbor with Proc Modeclus in Statistical Procedures

Nearest neighbor with Proc Modeclus

Samryn — Thu, 06 Oct 2011 16:16:35 GMT

I am trying to get a list of nearest neighbors for a dataset using Proc Modeclus. It appears to be working well, except there is a limit to the ID and neighbor value that I need to change. It appears to be limited to 5 characters, and I need more than that. Does anyone know a way to change this?

ods _all_ close;

proc modelus data=in1 dk=11 neighbor;

var xpos ypos;

ods output test1;

run;

in1 has 861,075 observations, and the listing I get gives the nearest neighbor if the observation values are less than or equal to 99,999, but uses ***** for values larger than that.

Thank you for your help.

Nearest neighbor with Proc Modeclus

art297 — Thu, 06 Oct 2011 18:08:08 GMT

Which output table is truncating the results? It may be a column that simply needs to be assigned a wider format via proc template.

Nearest neighbor with Proc Modeclus

Samryn — Thu, 06 Oct 2011 21:06:38 GMT

That is a great idea, but I don't know where the 5 character limit is being defined. I changed the template for Neighbor, and formatted the values to be 8 characters long, and so the format changed, but the length did not.

Nearest neighbor with Proc Modeclus

art297 — Thu, 06 Oct 2011 21:28:41 GMT

I would have thought the default of 8 characters would have been applied::

OUTLENGTH=n

OUTL=n

specifies the length of those output variables that are not copied from the input data set but are created by PROC MODECLUS.

The OUTLENGTH= option applies only to the following variables that appear in all of the output data sets: _K_, _DK_, _CK_, _R_, _DR_, _CR_, _CASCAD_, _METHOD_, _NJOIN_, and _LOCAL_.

The minimum value is 2 or 3, depending on the operating system. The maximum value is 8. The default value is 8.

Nearest neighbor with Proc Modeclus

Samryn — Thu, 06 Oct 2011 22:06:13 GMT

Ok, I think I found the problem. To define a length for the output, you have to define an ID variable. I defined the length of the ID variable in the data step just to make sure.

data in1;

set in1;

idv=_n_;

ods _all_ close;

proc modelus data=in1 dk=11 neighbor;

var xpos ypos;

id idv;

ods output test1;

run;

Thank you for your help Art.