Date:         Tue, 8 Feb 2005 13:18:10 +0100
Reply-To:     Magnus Källhager <magnus.kallhager@uc.se>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Magnus Källhager <magnus.kallhager@uc.se>
Subject:      Re: Sorting duplicate cases
Comments: To: Spousta Jan <JSpousta@CSAS.CZ>
In-Reply-To:  <CFCA22517B94B34C81F14B8DC44A39D55DCDFB@xi0sb001.cen.csin.cz>
Content-Type: text/plain; charset="ISO-8859-1"

Thanks Jan for your comment on this case.

I´m sorry if I was a bit short in the presentation of my problem. The main 
problem is that i havn´t got years for each firm but dates. Sure i can 
truncate the dates and get the years but still the "years" would not be 
correct since an annualreport period could be from six to 18 months. 
That´s why i would like to group them as the latest ones in one group and 
the second latest ones in another.

The way i do it now is to "identify duplicate cases" and save those away 
into another file and then start over again until i have gone thru the 
whole file.. It would be much easier with just a tag on each case telling 
me the order in group of duplicates.

Kindly,

Magnus Källhager




Spousta Jan <JSpousta@CSAS.CZ> 
Sänt av: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
2005-02-08 12:07
Sänd svar till
Spousta Jan <JSpousta@CSAS.CZ>


Till
SPSSX-L@LISTSERV.UGA.EDU
Kopia

Ärende
Re: Sorting duplicate cases






Hi Magnus,

I suppose that you have a variable YEAR - year of the annual report, and 
that every company has only one (or none) report in one year. Then simply 
compute

compute idyear = 2005 - year .

And you get how much years ago was the report. If you select idyear = 3, 

you will have 3 years old reports only.

If one or both of my premises do not hold, describe please your task and 
your data in more detail.

HTH

Jan



-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU]On Behalf Of 
Magnus Källhager
Sent: Tuesday, February 08, 2005 11:14 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Sorting duplicate cases


Hello List,

I have a set of data containing firms with their annual reports listed by
idnumbers for each company(duplicate cases). Now i´d like to identify the
duplicate cases and set casenumbers to them. Something like the $casenum
function but starting at each PrimaryFirst.

This would then make it possible to for example select "$casenumber 3"
wich will be the annual reports for all the companies three years ago.

Is this possible without to much hazzle??

// Magnus