Date: Mon, 20 Jun 2011 10:36:20 +0100
Reply-To: Garry Gelade <garry@business-analytic.co.uk>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Garry Gelade <garry@business-analytic.co.uk>
Subject: Re: Running in total
In-Reply-To: <1308490459708-4503507.post@n5.nabble.com>
Content-Type: text/plain; charset="us-ascii"
I'm not exactly sure what you want from looking at your example. However,
first of all I'd number your cases by something like:
COMPUTE const = 1.
COMPUTE casenum =$casenum.
SORT CASES by orgnr year casenum.
split file orgnr year.
create splitcasenum=csum(const).
(casenum is only there to ensure you don't inadvertantly screw up the
original case order)
Then calculate the maximum number of cases within each split (splitcasemax)
using AGGREGATE wih /BREAK = orgnr year using MODE = addvariables and
/splitcasemax = MAX(splitcasenum).
Then something like:
DO IF (splitcasemax - splitcasenum) > 6.
COMPUTE desire2 = 0.
ELSE.
COMPUTE desire2 = desire.
END IF.
You will then have the last 6 values in each split in desire2, and you can
calculate a running sum.
split file orgnr year.
create cumdesire=csum(desire2).
Hope this helps.
Garry Gelade
-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
DEBOER
Sent: 19 June 2011 14:34
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Running in total
Hi all, I have a problem to solve. I want to calculate the running total,
but
for only the last 6 cases (in points) with respect to orgnr and year.
I solved the cumulative variable as follow.
sort cases by orgnr year.
split file orgnr year.
create cumulative=csum(points).
execute.
The question is how can I create the desire variable so that we only
calculate the last six cases in the variable of points? Please see below at
the desire variable
<<thanks in advance.
orgnr points year cumulative desire variable(last 6)
1 1 2 004 1
1 1 2 004 2
1 -3 2 004 -1
1 -5 2 004 -6
1 5 2 004 -1
1 7 2 004 6
1 2 2 004 8 1
1 5 2 004 13 8
1 3 2 004 16 19
1 2 2 004 18 36
1 3 2 004 21 60
1 4 2 004 25 82
1 5 2 004 30 101
1 7 2 004 37 123
1 8 2 004 45 147
1 4 2 005 4
1 2 2 005 6
1 3 2 005 9
1 - 2 005 9
1 3 2 005 12
1 7 2 005 19
1 8 2 005 27 59
1 9 2 005 36 82
1 3 2 005 39 112
1 -4 2 005 35 142
1 -3 2 005 32 168
1 3 2 005 35 188
1 2 2 005 37 204
1 5 2 005 42 214
1 4 2 005 46 220
2 1 2 004 1
2 1 2 004 2
2 -3 2 004 -1
2 -5 2 004 -6
2 5 2 004 -1
2 5 2 004 4
2 2 2 004 6 -1
2 5 2 004 11 4
2 3 2 004 14 13
2 2 2 004 16 28
2 3 2 004 19 50
2 3 2 004 22 70
2 5 2 004 27 88
2 7 2 004 34 109
2 8 2 004 42 132
2 4 2 005 4
2 2 2 005 6
2 3 2 005 9
2 - 2 005 9
2 3 2 005 12
2 7 2 005 19
2 8 2 005 27 59
2 1 2 005 28 82
2 3 2 005 31 104
2 -4 2 005 27 126
2 -3 2 005 24 144
2 3 2 005 27 156
2 2 2 005 29 164
2 5 2 005 34 166
2 4 2 005 38 172
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Running-in-total-tp4503507p450
3507.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
