Date:         Thu, 22 Sep 2005 14:49:08 -0400
Reply-To:     Michael Wexler <mwexler@E-DIALOG.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Michael Wexler <mwexler@E-DIALOG.COM>
Subject:      Re: 2nd Attempt: Grouping Zip Codes
Content-Type: text/plain; charset="us-ascii"

Yet another reason to hope for an allegory to "proc sql" in SAS in some
future version of SPSS...

-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
Richard Ristow
Sent: Wednesday, September 21, 2005 1:53 PM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Re: 2nd Attempt: Grouping Zip Codes


At 09:40 AM 9/21/2005, Spousta Jan wrote:

>Only a few thoughts:
<<remarks on the inherent difficulty of the problem>>

>3) The question is then, whether SPSS is just the best tool to create
>such algorithm - another possibility is to use a general programming
>language. It can be better suited to go through the data in logically
>complicated ways. (SPSS and similar tools use the top-to-bottom way
>automatically - they just read/write all cases in one step...)

For an appropriate tool, I'd consider a relational DBMS that supports
SQL. It has syntax and semantics for many-to-many merges, and this
problem tends to need them. What do you think, Jan?

As Jan wrote, the problem is difficult. No tool can solve the
conceptual problems. The computational problems could manifest
themselves in SQL as straightforward, logically correct, solutions
needing storage and computation utterly beyond what's attainable.