Hi , You can use the merge also for the desired result . data want. merge a (in = a ) b (in = b) ; by id ; if a and b ; if xxx = yyy ; run; thanks
... View more
Hi , You can use the below set of code ,by moving the age2 before the output statements. data have; input sex$ age; datalines; F 14 M 15 ; data male female; set have; age2=age*1; if sex='M' then output male; else if sex='F' then output female; run;
... View more
Hi , You can use the below approach . data ranked ; set have ; if Consolidation_code = "C1" then rank = 1 ; else if Consolidation_code = "C2" then rank = 2 ; else if Consolidation_code = "C*" then rank =3 ; else if Consolidation_code = "U1" then rank = 4 ; else if Consolidation_code = "U2" then rank = 5 ; else if Consolidation_code = "U*" then rank = 6 ; else rank = 9 ; run; proc sort data = ranked ; by Consolidation_code rank ; run; data final ; set ranked; by Consolidation_code rank ; if first.Consolidation_code ; run; thanks
... View more
Hi , You can use the starta option with all option which will create a varialbe "selected"which have value 1 if the form 5% else 0 for left 90% . Proc sort data=have;by group;run; Proc surveyselect data=have ALL out=want samprate=5; strata group; run;
... View more
Hi , you can also use below code . data hh_prods; merge nodups_hhlevel_data (in=a) dups_hhlevel_data(in=b); by name1 ADDRESS zip account_id; if a and b then prod_code=1; else prod_code+1; run;
... View more
Hi Cherric , I have a another automated way which you can try, here you dont have to type name if the string have 1000 names. . data all; drop string; string = ' name1*name2*name3~name4*name5~name6 '; do until(word=' '); count+1; word = scan(string, count,'*~'); output; end; run; if you want it in horizontal way you can transpose the same
... View more
Hi , You can try the below approch ; data want ; set have ; if notdigit(acct) = 0 and lenght(acct) = 6 and 100000 =<acct >=999999 then derived = acct ; run;
... View more
Hi All , I have a question can anybody explain what the below part is doing and its requirement and why we have taken (-2 ). call symputx('tt1',put(intnx('day',today(),-2),yymmddn8.),G); Thanks
... View more
Hi , you can also use the first. last. approach as below . proc sort data = ORwfdTerm ; By Person_Number decending Termination_Date; run; data want ; set have ; By Person_Number decending Termination_Date;; if first.Person_Number ; run; thanks
... View more
Hi Zino , proc mean will work perfectly fine in this case . proc means data = have ; var 2007 2008 2008 2009 ; class id ; run; Hope this will help you
... View more
Hi Pankaj , i can think of a vague approach you can modify it accordingly ; we will use an array and multiple do loop for this ; part of code is as below ; array info (*) info_0 -- info_21 ; %do i = 0 to until (info_&i = "DIA") ; x= info_&i ; %end ; %do i = x to until info_&i in (D6D , D3A) ; dstart = 1 ; dend = 1 ; y = info&i ; end; and you can add loops as per requirement
... View more
Hi , We can also use the below approach . data have ; input date ; year = input(substr(strip(date),1,4),6.); month= input(substr(strip(date),5,2),4.); datalines; 196001 196002 196003 196004 ; run;
... View more
