If you want better visibility for your post it would be better to open a new thread and reference the old one with a link, especially since your topic is a lot broader. ... View more

I meant proc cluster; in SAS/STAT. Anyway,let's find out a way, an executable way. ... View more

Hi Reeza,      Thanks.I'm using SAS Enterprise Guide which installed on the server,and every time I run  %include "E:\TDDOWNLOAD\tagsets_9.1\excltags.tpl", it return an error that cann't store TAGSET,so if the template has to be saved someplace,I have to write the template to WORK.      So I'll set ODS PATH WORK.TMPL(update) SASUSER.TEMPLAT(update) SASHELP.TMPLMST(read) before %include.      Thanks very much. ... View more

Thanks Cynthia. I download excltags.tpl from http://support.sas.com/rnd/base/ods/odsmarkup/SAS 9.1 tagsets (.zip). And it works. Thanks. ... View more

FriedEgg, Thanks! I learn from yours and chang_y_chung's.]And find the way. data m01; input id score @@; cards; 1 12 1 4 1 20 1 109 2 189 2 18 2 99 3 12 3 18 3 146 4 27 4 99 5 80 5 100 5 180 6 101 6 700 7 12 8 99  ; run; data _NULL_; length has12 has18 hasge100 min1 min2 new_score 8.; if _N_=1 then do; declare hash h(ordered:"a");      h.definekey("id");      h.definedata("id","has12","has18","hasge100","min1","min2","new_score");      h.definedone();     end; set m01 end=last; if h.find() ne 0 then do; has12=0;has18=0;hasge100=0; min1=9999;min2=9999; new_score=0; h.add(); end; if h.find()=0 then Do; if score=12 then has12=1; if score=18 then has18=1; if score ge 100 then hasge100=1; /* h.replace();*/ max=max(score,max); min1=min(score,min1); if hasge100=1 then min2=min(score,min2); new_score=min1; if hasge100=1 then new_score=min2; if has18=1 then new_score=18; if has12=1 then new_score=12; if has12=1 and has18=1 then new_score=12; /* if has12=1 then new_score=12;*/ /* if has18=1 then new_score=18;*/ /* if has12=1 and has18=1 then new_score=12;*/ /* if hasge100=1 then new_score=min2;*/ /*  else new_score=min1;*/ h.replace(); end; if last then h.output(dataset:"m02"); run; proc print data=m02(keep=id new_score) noobs;run; ... View more

Yes, you can do clustering on a single variable. Suppose you know that there are two groups in your data and want to separate them automatically, you could use clustering to do that. Run the following example: /* Generate example data with 2 clusters in variable x */ data test; do x = 1,2,3,4,5,12,13,14; id = put(x,2.); output; end; run; /* Form all clusters */ proc cluster data=test outtree=tree method=centroid noprint; var x; id id; run; /* Isolate top 2 clusters */ proc tree data=tree out=clusters nclusters=2 noprint; run; /* Get quantiles */ proc rank data=test fraction out=quantiles; var x; ranks quantile; run; /* Assemble clusters and quantiles */ proc sql; select Q.id, Q.x, Q.quantile label="Quantile", C.cluster from clusters as C inner join quantiles as Q on C._NAME_=Q.id order by cluster, x; quit; Observations are assigned to the proper cluster. The quantiles are just like rescaled ranks. PG ... View more