@SASComm1 wrote: Much appreciated if one can show how to split the values into a single column in multiple rows. See below @Ksharp's sample script repurposed for a single value in multiple rows. data have; text='xx,ee,tt '; output; text='jj,ii,ii,kk'; output; run; data want; set have; do i=1 to countw(text,','); text2=scan(text,i,','); output; end; drop i; run; ... View more

data have; infile cards expandtabs; input x y ; cards; 0 4 0 4 0 4 0 4 0 4 16 4 0 4 0 4 0 4 0 4 0 4 0 4 16 4 0 4 0 4 0 4 0 4 0 4 0 4 16 4 0 4 ; data have; set have; n+1; if x>0 and y>0 then z=x/y; run; data key; set have(where=(z is not missing)) ; do k=n-3 to n; output; end; keep k z; run; data want; if _n_=1 then do; if 0 then set key; declare hash h(dataset:'key'); h.definekey('k'); h.definedata('z'); h.definedone(); end; set have; rc=h.find(key:n); drop rc k n; run; ... View more

Which method do you use? dcreate() function, or an external command? ... View more

You can also use "inner join" or "left join" clauses with proc sql ... View more

  This is just for fun. I wanted to experiment how a hash alternative would work at my college lab this afternoon.     @novinosrin If there was any uncertainty about your 'geek' status or SAS expertise consider it certain now 🙂 ... View more

Don't include the trailing spaces in your search pattern. where upper(c.product) like '%' || upper(trim(m.Med_product)) || '%' or upper(p.product) like '%' || upper(trim(c.Med_product)) || '%' Or better still just use the FIND() function. where find(c.product,m.Med_product,'it') or find(p.product,c.Med_product,'it') Also note that the UPPER() function is only available in SQL code. The normal SAS function of UPCASE() can be used anywhere in SAS code. ... View more

You can use _n_ and concatanate it with variabli. Data new; Set old; X=_n_; Y="var"||X||"_"||my_variable; Run; You can also use catx, ... View more

@Yavuz wrote: Is there a "period" column in your each dataset ? If there is no, then you must create one from the dataset name, before joining all datasets. If you dont do this you cannot compare data each other. Because all of them mixes.     There's the INDSNAME option that can be used in this situation, assuming SAS 9.3+. It will capture the name of the source file and then you can parse it to get the date from the file name. Here's a useless example of how that works:   data example; set sashelp.class sashelp.cars indsname=source; file_input = source; run;   @maryam1 you can use the methods I've illustrated in the link below, or Tom's code above which is a bit cleaner than mine for sure.    https://github.com/statgeek/SAS-Tutorials/blob/master/Import_all_files_one_type   1. Call the macro with  %list_files(C:\_localdata\temp, sas7bdat); This will create a data set called LIST that has the list of file names.  Then you can either proceed with the option from Tom's earlier post, use CALL EXECUTE to generate your Data Step combine statement or use those to assign a single libname to all folders.      ... View more

Data data2; Set data1; If (month>t) then output data2; Run; ... View more