Basically you could use Proc transpose with ID statement for PAST_DUE variable, to transform it into wide format. However, once transposed data values in the PAST_DUE column get converted into Variable name in compliance with sas variable naming  conventions (starting either with letters or underscore, containing letter, underscore or numbers, and not longer than 32 character). This will transform spaces or special characters into underscore, truncate long strings if more than 32 characters.  Besides, the output will not look like as you shown above as your location values are different. There will be one record per distinct location.  eg. try this and see; proc transpose data= have out=want; by location; var portfolio_amt; id past_due; run; ... View more

Hi,  Your code seems to return a valid value which is missing when pending='yes';  ... View more

If you only need lines on the top and the bottom of your report, try to group also grpxsort1 variable, and add another compute block to add a line before, that's something like this: data have; infile cards dsd; length ae $25; input grp1 grp2 ae $; cards; 1, 0, List of AE by group 1, 1, Group 1 Disorders 1, 1, Nausea 1, 1, Anaemia 1, 1, Vomiting 1, 2, Group 2 Disorders 1, 2, Pyrexia 1, 2, Fatigue 1, 2, Malaise ; proc print;run; ods rtf file= "&lib\have.rtf"; proc report data=have ; column grp1 grp2 ae; define grp1/ group noprint; define grp2/ group noprint; define ae/ 'Adverse Events' display; compute before grp1; line @1 ""; endcomp; compute after grp2; line @1 ""; endcomp; run; ods rtf close;   Output:  Style attributes could be added to apply a desired indentation and format.     ... View more

You need to GROUP your grpxsort2 variable in the define statement.  define grpxsort2/ group noprint;     ... View more

You should use style (header)= attributes in the define statements (which is individual) and remove justifications from the proc report statement (which is general).   proc report data= dsin nowd missing split = '~' style(report)={frame=box} style(header)={font=('Times New Roman', 8 pt) background=lightgrey} style(column)={font=('Times New Roman', 8 pt) }; column label sample; define label / display "This is a header" style(header)= {just=left} style(column)={just=left cellwidth = 12.0 cm asis=on}; define sample / display "sample" style(header)= {just=center} style(column)={just=center cellwidth = 3 cm}; run;   ... View more

Hi,    Use DATA step to read  a csv file. You cannot control over the format in PROC IMPORT.  ... View more

Hi,  If your date variable type is numeric then it should sort correctly, regardless the format used.  But if your variable type is character then you need to convert it into numeric, using INPUT() function and assigning the desired format (eg. DATE9. );  Example: data want; set have; date1= input(date, mmddyy10.); format date1 mmddyy10.; run;    ... View more

Hi,  What does the warning message say?  And also you are not using .xml extension in ENAME. I believe tagstets.excelxp is not compatible with.xls extension which was used in the program.  ... View more

Try this: data ex1; set sashelp.class ; where strip(name) like '%a_' ; proc print;run; I'm assuming "Barabara" has a trailing blank, and the wildcard (_) picking this blank as a single character, so the last letter 'a' is being considered as the second letter backwards.  ... View more

Hi it would be great to provide a portion of data, even dummy, for testing. There was a similar post https://communities.sas.com/t5/SAS-Studio/MERGE-statement-has-more-than-one-data-set-with-repeats-of-BY/td-p/449980. I would also try the syntax : if A and B and C and D and E; in lieu of : if in (a, b, c, d, e); ... View more

You might also consider Proc Report as an alternative where STYLE attributes forces excel to print values in desired format. Eg. the below code prints leading zeros: proc report data=test; column _all_; define x/display style(column)={tagattr='format:000'}; run; ... View more

You can create dynamic statement using COUNTC function and %do block. However, the code should be in macro definition: Try this: %macro print; proc print data=HAVE; where &value_1 IN (%do i= 1 %to (%sysfunc(countc(&value_2, |))+1); "%scan(&value_2, &i, |)" %end;); run; %mend; %print;     ... View more

While not sure about whether IN operator could accept the PIPE as a delimiter,  I suggest using %scan function to separate pipe delimited values into individual values inside the parenthesis.  If macrovar has only 2 values then the following code would separate them into individual values to process using IN operator: where &value_1 IN ("%scan(&value_2, 1, |)", "%scan(&value_2, 2, |)"); ... View more