>Is there any way to model with variable ne (^=). I did try but it is not working. Yes. proc mixed data=spring07(where crop ne 'corn' ) ...... >Phos(weighted0-40) = (5*Phos1 + 5*Phos2 + 10*Phos3 + 20*Phos4)/40. You can just make hard code to get it. data sprint0; Phos_weighted= (5*Phos1 + 5*Phos2 + 10*Phos3 + 20*Phos4)/40. BTW, the best way to get your answer is to give your origin data and what your output looks like. Ksharp ... View more

May be you need macro variable. Long time no see. @data _null_; BTW,You also can search dictionary table " dictionary.tables " to find whether the dataset you needed is exist.then set a Return Code to it customsizely . [pre] %let file_name=sashelp.class; %let file_exist= %sysfunc(exist(&file_name)); %put NOTE: 1 is exist , 0 is non-exist: &file_exist; DATA LNX_USERS; length filename $256; filename = "&file_name"; if &file_exist then do; INFILE DUMMY DELIMITER=',' DSD FIRSTOBS=2 end=eof filevar=filename; do while(not eof); INPUT GUEST :$8. ... KBSWPCAD :8.; LNXDATE = TODAY()-1; output; end; end; stop; run; [/pre] Ksharp Message was edited by: Ksharp ... View more

Hi. If you have understand '@' and '@@' totally. then code will be more simply! [pre] data people; infile datalines delimiter=',' truncover; input name $ score @; do until ( missing(score) ); output; input score @; end; input; /* release the input hold */ datalines; David,10,3,20 Mary,31,2 Jason,1 ; run; proc print noobs;run; [/pre] Ksharp ... View more

[pre] data have; input id $ month : date9. activity ; format month date9.; cards; 0001 01Jan09 4 0001 01Mar09 3 0001 01Jul09 1 0002 01Feb09 2 0002 01Apr09 2 0003 01Dec10 3 ; run; proc sort data=have; by id month; run; %let start=01jan09; %let end=01dec09; data temp; set have; _month="&start"d; if id ne lag(id) then do; do i=1 to intck('month',"&start"d,"&end"d)+1; _id=id; _activity=0; output; _month=intnx('month',_month,1,'sameday'); end; end; format _month date9.; keep _id _month _activity; run; proc sort data=temp(rename=(_id=id _month=month _activity=activity)); by id month; run; data want; merge temp have; by id month; run; [/pre] Ksharp ... View more

You are right. Regards. Ksharp ... View more

No .Art T I do not think so. Data step also can do Cartestian Join , just need more code than sql. I am just coding in data step to simulate proc sql 's Cartestian Product. And at the opposite, There are some case that only data step can do, sql cann't. And for the sake of large dataset (such as 100G), Proc sql is almost not to run,now it is show time for SAS data step. Ksharp Message was edited by: Ksharp ... View more

Hi. Are you to try to use ' notsorted' option in proc format for customizing order of value. [pre] proc format; value fmt(notsorted) 1,2 = 'GH' 3,4 ='DK' 5,6='SV' 7,8='AF' ; value _fmt(notsorted) 1,2 = 'DK' 3,4 ='SV' 5,6='GH' 7,8='AF' ; data temp; input a b; format a _fmt.; format b fmt.; cards; 1 5 2 8 3 6 2 1 8 3 7 2 6 4 ; run; proc tabulate data=temp; class a b; table a,b; run; [/pre] Ksharp ... View more

Hi. I try it. In fact, SAS will use your illegal variable name as label.And variable is actually converted successful. If you add another options ' options nolabel ; ' then you will find SAS has already convert varname as the rule of V7. Or use proc contents to see it deeply. Ksharp ... View more

Hi. It looks like you are using proc sql 's Pass-Through capability. Do you try to use libname statement? Ksharp ... View more

Hi.Art T. It looks like that it is time for me to find more in documentation. Ksharp ... View more

WOW.Art T I bet that Art Carpenter will definitely like your code. My code is only another choice. [pre] data have; input ProvID Company $ Rate Payment; cards; 1234 CompA 20.2 100.00 1234 CompA 30.5 200.00 1234 CompA 40.7 300.00 1234 CompB 20.2 150.00 1234 CompB 50.5 250.00 ; run; * Get magic number; proc freq data=have noprint order=freq ; tables provid*company /list out=count ; run; data _null_; set count; if _n_ = 1 then do; call symputx('count',count); stop; end; run; * transpose to wide; proc sort data=have; by provid company; run; data result(drop=rate payment i); set have; by provid company; array r{*} rate1-rate&count; array p{*} payment1-payment&count; retain rate: payment:; i+1; r{i} = rate; p{i} =payment; if last.company then do; output; i=0; call missing(of r{*} p{*}); end; run; proc print ;run; [/pre] Ksharp ... View more

Yes. There is a system option ' options validvarname=any ' can use any special character for variable name (only available for Base and Stat). If you do not want, can close it or change it. Ksharp ... View more

Agree!. Each of Data-step and Proc-sql has its virtues respectively,then If have skill of both ,you will be a perfect SAS programmer. Ksharp ... View more

Set them missing value Or using $varying. format. Ksharp ... View more

[pre] data temp; length _name $ 20; set sashelp.class; if _n_ in ( 2 4 6 8 10 ) then _name=cat(' ',name); else _name=name; run; ods pdf file='c:\title.pdf' style=sasweb; ods escapechar='~'; title j=l '~S={asis=on background=darkblue foreground=white}' ' font is white '; proc report data=temp nowd ; column _name name weight; define _name /display ' name of' style( header column)={asis=on}; define name /display; define weight/display; run; ods pdf close; [/pre] Ksharp ... View more