Please post the result dataset as data step, so we can use your data for analysis. The number of obs shown in the screenshot is sufficient. We don't anything more, but those two variables. ... View more

I have merged your posts. Please don't double-post question. ... View more

Proc contents prints metadata, the where-option has no impact on the metadata of the processed dataset, there are still 180 obs in your dataset. ... View more

A solution depends on the os of the sas server. On a windows server i would use something like the solution posted on https://stackoverflow.com/questions/17829785/delete-files-older-than-15-days-using-powershell.   ... View more

As the column data have repeated values i have used Proc SQL -Full join but i am not able to sort by date.Please anyone help me. Please post the code, the log and explain why do you think that sorting by date doesn't work. ... View more

Proper dates would make this easier.   With your current data: proc transpose data=work.t1 name=source out=work.transposed(rename=(col1=month)); by ID; var Month:; run; proc sort data=work.transposed out=work.sorted; by id month; run; data work.want; merge work.score_tbl_all work.sorted; by id month; if not missing(source); source = propcase(source); run; Not the expected shape, but long tables are almost always better to work with. ... View more

Why do you want to distribute the obs in c4? Creating multiple datasets is not necessary and sometimes it will even increase the complexity of following steps. ... View more

And what should happen if exactly 10 answers are correct? Posting data in usable form helps us to provide tested code. ... View more

You have checked the spelling? The directory is really named INPUT, not "input"? The filename has only uppercase letters? ... View more

Var7 needs to be unique, so that by var7 can be used in proc report. That should create a single table for each value. I still don't know how to include the variables label in the output of the byline, and, because i haven't used compute-blocks in proc report, i still have no idea how they could be defined to print var5 and var6 as requested.   Maybe @Cynthia_sas has an idea. ... View more

I hoped that var7 has a unique value in each observation, so that proc report could be used, but with the data shown, i don't think that proc report could be used, but i am not super-familiar with the proc.   Report Writing Interface could be used, but you will have to write statement for anything you want to see and even for some things you don't want to see. Here's on possible approach: %let dummy_header = %str(backgroundcolor=colors('headerbg') color=colors('titlefg') ); data _null_; set work.have end=jobDone; length _dummy $ 200; if _n_ = 1 then do; declare odsout rwi(); end; rwi.table_start(); _dummy = catx(': ', vlabel(var7), var7); rwi.row_start(type: 'h'); rwi.format_cell(data: _dummy, inhibit: 'x', column_span: 4, just: 'L'); rwi.row_end(); rwi.row_start(type: 'h'); rwi.format_cell(data: vlabel(var1)); rwi.format_cell(data: vlabel(var2)); rwi.format_cell(data: vlabel(var3)); rwi.format_cell(data: vlabel(var4)); rwi.row_end(); rwi.row_start(); rwi.format_cell(data: var1, just: 'R'); rwi.format_cell(data: var2, just: 'R'); rwi.format_cell(data: var3, just: 'R'); rwi.format_cell(data: var4, just: 'R'); rwi.row_end(); rwi.row_start(); rwi.format_cell(data: vlabel(var5), column_span: 4, just: 'L', style_attr: "&dummy_header."); rwi.row_end(); rwi.row_start(); rwi.format_cell(data: var5, column_span: 4, just: 'L'); rwi.row_end(); rwi.row_start(); rwi.format_cell(data: vlabel(var6), column_span: 4, just: 'L', style_attr: "&dummy_header."); rwi.row_end(); rwi.row_start(); rwi.format_cell(data: var6, column_span: 4, just: 'L'); rwi.row_end(); rwi.table_end(); run; ... View more

Hardly possible to suggest something without knowing how the data looks like. So please post an excerpt of the data you have in usable form: a data step using datalines statement.   ... View more