Date:         Thu, 28 Oct 2010 13:41:43 -0400
Reply-To:     Suzanne McCoy <Suzanne.McCoy@CATALINAMARKETING.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Suzanne McCoy <Suzanne.McCoy@CATALINAMARKETING.COM>
Subject:      Re: rolling up counts in PROC FREQ based regex patterns
Comments: To: "White, Svend A." <SvendW@HEALTH.OK.GOV>
In-Reply-To:  <54E01EE17F635448A5CE0C2253E23FF9029421F3@CONSTANCE.DOH.OK>
Content-Type: text/plain; charset="us-ascii"

This is where DataFlux comes to the rescue if you have it in your software bundle.

-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of White, Svend A.
Sent: Thursday, October 28, 2010 1:32 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: rolling up counts in PROC FREQ based regex patterns

I'm running frequencies on addresses to identify the most common
patterns of invalid street addresses.

What I'm wondering is this: Is there a way to instruct PROC FREQ to roll
up counts based on regular expression patterns (e.g., combine the counts
for "Rural Route 1" and "Rural Route 10", using a Perl regex like "Rural
Route \d+").

Thanks.

Svend