data have; input loan_num sales returns; datalines; 112 12 0 113 11 10 114 9 5 ;run; data have2; set have; diff=sales-returns; run; diff=sales-returns works in this example.  In my actual program  I created a variable named diff.  I subtract sales and returns.  Both are defined as numeric.  It calculates correctly if both numbers are >0, however if one of the numbers is a 0 the diff is shown as 0 which is incorrect.  My question is would there be a different method of subtracting the two variables.  Intck is appropriate for dates, not two numerics as shown here. ... View more

data test1; infile datalines; input setup_date date9. Division $ cnt; return; datalines; 1Jun2018 Central 4 1May2018 Central 5 1Apr2019 Atlantic 3 1Feb2019 Atlantic 11 1Jan2017 Pacific 11 1Mar2019 Central 11 ;run; data mydata; length year2 $12.; set test2; year1 = year(setup_date);/*returns numeric*/ monthsum = month(setup_date);/*returns numeric*/ year2 = put(year1,4.);/*change to character*/ Month1 = put(setup_date,monname3.);/*change to character*/ format setup_date date9.; run; proc sort data=mydata;by year1 descending cnt;run;   Produces this output year2 setup_date Division cnt year1 monthsum Month1 2017 01Jan2017 Pacific 11 2017 1 Jan 2018 01May2018 Central 5 2018 5 May 2018 01Jun2018 Central 4 2018 6 Jun 2019 01Feb2019 Atlantic 11 2019 2 Feb 2019 01Mar2019 Central 11 2019 3 Mar 2019 01Apr2019 Atlantic 3 2019 4 Apr   proc report data = mydata nowd style(summary)=Header; column year2 monthsum Month1 cnt rtot; define year2 /group ; define monthsum /group noprint order = data; /*will sort the month in month order*/ define Month1 /across order = data ; define cnt /sum f=comma6.; define RTot /computed f=comma16. "Row Totals"; compute RTot; RTot=sum(cnt.sum);/*row_tot*/ endcomp; compute before; sum=cnt.sum; endcomp; compute after; year2='Grand Total'; endcomp; break after year2 / ol summarize skip;/*total by group*/ rbreak after /summarize; run; Produces this proc report   Month1   year2 Jan May Jun Feb Mar Apr cnt Row Totals 2017 1 . . . . . 11 11 2017 1 . . . . . 11 11 2018 . 1 . . . . 5 5   . . 1 . . . 4 4 2018 . 1 1 . . . 9 9 2019 . . . 1 . . 11 11   . . . . 1 . 11 11   . . . . . 1 3 3 2019 . . . 1 1 1 25 25 Grand Total 1 1 1 1 1 1 45 45   desired output is to move the year from row the column and segment the month by the individual year Desired outout                                   2017 2018 2019 CNT     Jan May Jun Feb Mar Apr cnt Row Totals   1 . . . . . 11 11   1 . . . . . 11 11   . 1 . . . . 5 5   . . 1 . . . 4 4   . 1 1 . . . 9 9   . . . 1 . . 11 11   . . . . 1 . 11 11   . . . . . 1 3 3   . . . 1 1 1 25 25   1 1 1 1 1 1 45 45 ... View more

data _null_; call symput ('timenow',put (time(),time.)); if &timenow >= 17:30:00 and &timenow < 17:35:00 then %let reportname= LSR2_Down_Report_4;else if &timenow >= 20:30:00 and &timenow < 24:35:00 then %let reportname= LSR2_Down_Report_6; run; %put &timenow &reportname;   I am attempting to assign a reportname based on the time of day. When I run the code I get the following error SYMBOLGEN: Macro variable TIMENOW resolves to 17:40:48 49 if &timenow >= 17:30:00 and &timenow < 17:35:00 then SYMBOLGEN: Macro variable TIMENOW resolves to 17:40:48 NOTE: Line generated by the macro variable "TIMENOW". 49 17:40:48 _ 388 76 ERROR 388-185: Expecting an arithmetic operator. ERROR 76-322: Syntax error, statement will be ignored.   ... View more

data have; length ln Status $25. ; input ln Status Month date9. Curr Due_030 tot; return; datalines; 122 Current 01Nov2019 1 0 100 122 Current 30Nov2019 1 0 1 122 Due_030 01Dec2019 0 0 5 122 Due_030 31Dec2019 0 1 2 ;run;   The last column is called tot.  Is there a way to repeat the top number (in this case 100) on all rows with a column name tot2 I want to get the percentage of tot by tot2 (ie 100/100, 1/100, 5/100) ... View more

data have; length Status $25.; input Status cnt; return; datalines; Tot_pop 10 Curr 5 _30_day 1 Paid 4 ;run;   Status cnt Tot_pop 10 Curr 5 _30_day 1 Paid 4   Is there a way to calculate the percentage change using Tot_Pop as the base line.  Idea is to divide Curr/Tot_pop and show results on Curr row, then _30_day/tot_pop and show on _30_day row? ... View more

data test; length Category Month $20. ; input Category $ Month $ Curr Due_30 Paid; datalines; Current Jan19 4 0 0 Paid Jan19 0 0 2 Due_30 Feb19 0 7 0 Due_60 Feb19 0 2 0 ;run; output is Category Month Curr Due_30 Paid Current Jan19 4 0 0 Paid Jan19 0 0 2 Due_30 Feb19 0 7 0 Due_60 Feb19 0 2 0   proc format; value $catfmt 1='Current' 2='Due_30' 3='Paid' ; value $monthfmt 1='Jan19' 2 ='Feb19' 3 ='Mar19' ; value nu run; proc tabulate data=test ; VAR Curr Due_30 Paid; CLASS Month / order=internal /*ORDER=UNFORMATTED MISSING;*/; CLASS Category / order=internal /*ORDER=UNFORMATTED MISSING;*/; TABLE /* ROW Statement */ Category *(Curr*f=number8.0 Due_30*f=number8.0 Paid*f=number8.0 ) all = 'Total' *(Curr*f=number8.0 Due_30*f=number8.0 Paid*f=number8.0 ), /* COLUMN Statement */ Month *( Sum={LABEL=""} ) all = 'Total' *( Sum={LABEL=""} ) ; ; format Category $catfmt. Month $monthfmt. ; RUN;   1. Despite using proc format to arrange the months in date order, (ie Jan,Feb) the report uses alpha order so Feb comes first 2. I want to should a %roll change based on each month.  For example in Jan19 there were 4 current accounts.  2 of the accounts paid off so I want to show a 50% change.  Since they are different statistics they show in different columns.  Therfore I can not use the lag function so can I address this in the proc tab ... View more

 data have; length ln Status $25. ; input ln Status Month date9. Curr Due_030; return; datalines;   122 Current 01Nov2019 1 0 122 Current 30Nov2019 1 0 122 Current 01Dec2019 1 0 122 Due_030 31Dec2019 0 1 ;run; data have2; set have; format month date9.; run; Produces this output ln Status Month Curr Due_030 122 Current 01Nov2019 1 0 122 Current 30Nov2019 1 0 122 Current 01Dec2019 1 0 122 Due_030 31Dec2019 0 1 proc transpose data=have2 out=have3 let; id month; by Status; ;run; Produces this output from transpose Status _NAME_ 01NOV2019 30NOV2019 01DEC2019 31DEC2019 Current Curr 1 1 1   Current Due_030 0 0 0   Due_030 Curr       0 Due_030 Due_030       1 My issue is I need to be able to compare the beginning status with the ending status by month and produce stats As an example for have 2 ln 122 began as current for Dec1 then became delinquent by Dec31.  Therefore the percentage roll from current to delinquent would be 100%.  I thought using proc transpose would allow me to analyze horizontal instead of vertical however look what transpose does to the Due_30.  Is there a way to do the analysis in a "roll" fashion while looking at changes in status for the 1 month time period as shown here.  I am using this as an example.  I also enclose a mockup of the actual report I need to re-create   ... View more

data _NULL_; CALL SYMPUTx('mth1',put(intnx('year',today(),-1,'b'),monyy5.),'g') ; CALL SYMPUTx('mth12',put(intnx('year',today(),-1,'e'),monyy5.),'g') ; run; Output is JAN19 and Dec19 respectively   I am attempting to get remaining months exactly 1 full year ago starting with Jan19 How would I be able to use the same logic for say FEB19, MAR19...etc.   ... View more

data have; length ctr Status $25.; input ctr cnt cnt2 Status; return; datalines;   4544 2 1 Paid 4544 3 1 Paid  /*suppress the name paid but keep record*/ 4544 4 1 Paid 4544 5 1 Paid 4465 10 1 4465 5 1 4475 1 0 4475 2 0 Paid 4475 3 1 Paid /*suppress the name paid but keep record*/ 4476 5 2 4476 6 1 Paid ; run; Is there a way to keep all record however suppress the name "Paid" after it shows the first time? So in a series per ctr, if the status is "Paid", I want to see it displayed on the first instance only but keep all record.   ... View more

I began with the following data _null_ to attempt to get todays date in yymmdd10  (ie '2020-02-16') DATA _NULL_; CALL SYMPUTx('DEKEY', "'"|| PUT(INTNX('DAY',&sysdate9,-0,'END'),yymmddd10.) ||"'") ; RUN; %put &dekey;   I am attempting to get the first and last day of last year using the sysdate function (ie '2019-01-01'  and then a separate data _null for '2019-12-31') ... View more