example demonstrates how changing a default penalty affects the placement
of labels. The goal is to produce a plot that has labels that do not detract
from how the points are scattered.
options nodate pageno=1 linesize=120 pagesize=37;
proc plot data=census;
plot density*crimerate=state $ state /
placement=(h=100 to 10 by alt * s=left right)
haxis=0 to 13000 by 1000 vaxis=by 100;
title 'A Plot of Population Density and Crime Rates';
Copyright 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.