The DQSCHEME Procedure |
This example generates an analysis of the STATE variable in the VENDORS data set.
Note: You do not have to create a scheme to generate the analysis data set.
Note: The locale ENUSA is assumed to have been loaded into memory as part of the locale list.
/* Create the input data set. */ data vendors; input city $char16. state $char22. company $char34.; datalines; Detroit MI Ford Motor Dallas Texas Wal-mart Inc. Washington District of Columbia Federal Reserve Bank SanJose CA Wal mart New York New York Ernst & Young Virginia Bch VA TRW INC - Space Defense Dallas TX Walmart Corp. San Francisco California The Jackson Data Corp. New York NY Ernst & Young Washington DC Federal Reserve Bank 12th District New York N.Y. Ernst & Young San Francisco CA Jackson Data Corporation Atlanta GA Farmers Insurance Group RTP NC Kaiser Permanente New York NY Ernest and Young Virginia Beach VIRGINIA TRW Space & Defense Detroit Michigan Ford Motor Company San Jose CA Jackson Data Corp Washington District of Columbia Federal Reserve Bank Atlanta GEORGIA Target ; run; /* Create the analysis data set. */ proc dqscheme data=vendors; create analysis=a_state matchdef='State (Scheme Build)' var=state locale='ENUSA'; run; /* Print the analysis data set. */ title 'Analysis of state name variations'; proc print data=a_state; run;
For each value of the STATE variable, the analysis data set WORK.A_STATE shows the number of occurrences and the associated cluster number. Variables that are not clustered with any other values have a blank value for the cluster number.
Note: This example is available in the SAS Sample Library under the name DQANALYZ.
Copyright © 2010 by SAS Institute Inc., Cary, NC, USA. All rights reserved.