Cut-and-paste then run the syntax at the end of this message.
Does this do what you want? If not please restate your question in more
detail.
Hope this helps.
Art
[log in to unmask]
Social Research Consultants
University Park, MD USA
(301) 864-5570
new file.
* this program generates 20 cases in each of 2 groups.
* to adapt it to your needs you can
* 1) make more loop sets for more groups
* 2) change the 20 to whatever you need for each group
* 3) change the vector of probabilities for each group
* 4) change the number of x values
* 5) change the seed to get different random numbers.
input program.
set seed = 6092003.
loop #i = 1 to 20.
compute grp =1.
vector x(9,f1).
do repeat lx =x1 to x9
/p = .6 .7 .2 .3 .4 .7 .8 .6 .9.
compute lx = rv.bernoulli(p).
end repeat.
end case.
end loop.
loop #i = 1 to 20.
compute grp =2.
do repeat lx =x1 to x9
/p = .3 .3 .6 .5 .6 .9 .4 .4 .4.
compute lx = rv.bernoulli(p).
end repeat.
end case.
end loop.
end file.
end input program.
formats grp (f1).
execute.
QUICK CLUSTER
x1 x2 x3 x4 x5 x6 x7 x8 x9
/MISSING=LISTWISE
/CRITERIA= CLUSTER(2) MXITER(10) CONVERGE(0)
/METHOD=KMEANS(NOUPDATE)
/SAVE CLUSTER DISTANCE
/PRINT INITIAL.
CROSSTABS
/TABLES=qcl_1 BY grp
/FORMAT= AVALUE TABLES
/CELLS= COUNT .
Hédia Mhiri Sellami wrote:
> ----- Original Message -----
> From: Hédia Mhiri Sellami
> To: [log in to unmask]
> Sent: Monday, June 09, 2003 12:00 PM
> Subject: datasets_Urgent
>
>
>
> Dear Member;
>
> It is true that I can find many URL adresses from which I can download data set but what I'm looking for is not only data but also a proposition of an associated partition.
> I will take data to submit it to the SAS procedure FASTCLUS and compare its result with the partition dowloaded.
> In fact I'm working on a new criteria to compare partition and I need to test my criteria.
> I would like to ask you, if possible, to help me finding adress in which I can download data and an associated partition as soon as possible because of the deadline of call for paper to which I would like to submit my paper.
> Best regards
>
|