proc sql; create table test as select a.*,b.eps,b.date as date_event format=monyy6. from have_base as a, have_event as b where intnx('month',a.date, -12)<=b.date<=a.date; quit; ... View more

I was thinking about call execute as below, but it is not resolving the num wih a+b values. Any thoughts from experts would be appreciated   data t1; input index $ num $ num2 $; datalines; var1 a+b b var2 b b+c run; data t2; input a b c; datalines; 1 2 3 4 5 6 run; proc transpose data=t1 out=test; id index; var num num2; run; data want; merge t1(in=x) t2(in=y); run; data want2; set want; call execute('sum='||num||';'); run;   ... View more

did you try to increase the width of N_50mg N_100mg N_200mg in a way that these variable will fit in one page and next N_400mg N_600mg N_800mg will go to the next page and paran and _name_ are in the id so they will be displayed in both the pages. for the other question I think we need to use macro variables which will conditionally check the missing variables and drop them. ... View more

Please try this untested code , i used the compute blocks.   Also to get the variable N_50mg N_100mg N_200mg width incresed in such a way that these variable will fit in one page and next N_400mg N_600mg N_800mg will go to the next page and paran and _name_ are in the id so they will be displayed in both the pages.   /*this is for the line size, it could be change as per your requirement;*/ %let ls=68; proc report data = all4 nowd split = "|" style(report)=[asis=on] nowindows ; column avisit paran _NAME_ N_50mg N_100mg N_200mg N_400mg N_600mg N_800mg ; define avisit/ order 'analyte avisit'; define paran /id order "PK Parameter" format = sch.; define _NAME_/ id ""; define N_50mg/ display ; define N_100mg/ display ; define N_200mg/ display ; define N_400mg/ display; define N_600mg / display ; define N_800mg / display ; break after avisit / page; compute before _page_/ left; line 'analyte avisit: ' avisit; line &ls.*'_'; line ' '; endcomp; run; .  ... View more

    proc sgplot data=insurance; reg x=size y=month / group=stock; run; ... View more

First tranpose the ref dataset and then merge with source via proc sql as below to get the expected output     proc transpose data=ref out=trans; var date2; run; proc sql; create table want as select a.date, case when a.date>=b.col1 and a.date <=b.col2 then b.col1 else b.col2 end as date2 format=date9. from source as a , trans as b; quit; ... View more

proc sort data=have; by cust_id year descending customer_age month; run; data want; set have; by cust_id year descending customer_age month; retain new_age; if first.cust_id then new_age =.; if customer_age ne . then new_age = customer_age; customer_age=coalesce(customer_age,new_age); if customer_age^=.; drop new_age; run; proc sort data=want; by cust_id year month; run; ... View more

Thank you, I responded to the other thread. Please check. If the solution helped, could you please mark as answered. So it will help others if they have a similar query. ... View more

try the sum function   Data B; Set A; if missing(E) then SixVar=sum(0.2099*A, 0.1871*B, 0.1458*C, 0.1955*D, 0.2617*E); else if missing(E) then SixEFAVar=sum(0.2291*A ,0.1894*B , 0.1688*C, 0.1765*D, 0.2362*E); else if missing(C) then SixEFAVar=sum(0.2380*A, 0.1968*B, 0.1366*E, 0.1833*D, 0.2453*E); else if missing(D) then SixEFAVar=sum(0.2398*A,0.1983*B, 0.1768*C, 0.1377*E, 0.2473*E); else if missing(B) then SixEFAVar=sum(0.2432*A, 0.1792*C, 0.1396*E, 0.1873*D, 0.2507*E); else if missing(E) then SixEFAVar=sum(0.2558*A,0.2116*B, 0.1886*C, 0.1469*E, 0.1971*D); else SixEFAVar=sum(0.2024*A, 0.1674*B, 0.1492*C, 0.1163*E, 0.1559*D, 0.2087*E); Run; ... View more