@Tom wrote: @andreas_lds wrote: ... ... new_policy_number = compress(transtrn(upcase(policy_number), '_', '')); ...    Did you mean: new_policy_number = compress(upcase(policy_number), '_ '); In other words: remove the underscores and the spaces. Well, yes ..😟 ... View more

Some corrections: The word "manual" is part of "manualentry" and "cedemanual", so if "manual" is removed first, "cede" and "entry" would stay in the string. Using "_all_" in the array statement is not what you want, as it includes "policy_number" also. Upcase and removing blanks and underscores was moved before the loop. Using "policy_number" in the loops restores already removed words. Changed from tranwrd to transtrn (see docs) and added trim to get rid of trailing blanks. data want; set test_set; length new_policy_string $ 100; array words[4] $ 15 _temporary_ ('MANUALENTRY', 'CEDEDMANUAL', 'MANUALCEDED', 'MANUAL'); new_policy_number = compress(transtrn(upcase(policy_number), '_', '')); do i = 1 to dim(words); new_policy_number = transtrn(new_policy_number, trim(words[i]), ''); end; drop i; run; ... View more

I don't know any function that can do this, would be interesting to maintain such a function, which, of course, needs to support multiple languages.   Maybe such a function exists in a special sas-component. Maybe you can find a list of nouns in the www.   If you have such a list, the program could have the following steps: read the list create an informat (see code below) check each word data nouns_fmt; set list(rename=(noun = start)) end=jobDone; retain FmtName 'IsNoun' Type 'i' Hlo 'U' Label 1; Start = upcase(start); output; if jobDone then do; HLO = 'O'; Label = 0; output; end; run; proc format cntlin= nouns_fmt; run; ... View more

Thanks for posting data in usable form.   I know this is an exam task, but: The need to use a merge is artificially generated by creating multiple datasets. Proc transpose seems to be better suited to solve the problem. Or have failed to see something?   proc sort data=examEX4; by id KneeNumber; run; proc transpose data=examEX4 out=t_want prefix=score_knee name= Visit ; by id; var ScorePre Score1Day Score1Week Score1Month; run;   ... View more

Please don't double post questions. ... View more

Please post the data you have in usable form: a data step using datalines.   Without "some customers are reserved to managerX", proc rank could do the job, but i think some additional steps are required to fulfil the requirement. ... View more

I would combine all B datasets and add the year, the dataset will then look like your final A dataset. Then you can do something like:   data want_a; set a; if _n_ = 1 then do; declare hash h(dataset: "big_b"); h.defineKey("id", "year"); h.defineData("quantity"); h.defineDone(); end; rc = h.find(); run; ... View more

sashelp.vtable has the size of all datasets available in the session. Try:   data _null_; set sashelp.vtable; where libname = 'LIB' and memname = 'XYZ'; memsize = input(getOption('memsize'), best.); if filesize < memsize then do; call execute('sasfile lib.xyz open;'); end; run; ... View more

@neo_noone wrote: I am quite a novice in programming and I kind need your help regarding an issue I notice. I have a table: date ID secondary D E date1 b1 c1 d1 e1 date1 b c d e date3 b2 c1 d3 e3 date4 b2 c2 d4 e4 Id and secondary should be unique and always matching. But I notice that they are not. It's either correct the ID or the secondary ID. I want to sum all the D column per unique ID. Thanks you all for reading and if you have any inputs would be very helpful   The table has zero numeric variables (except for date, maybe), so there seems to be nothing that can be summed. Please post the data you have in usable form (a data step using datalines) and show the expected output for that dataset. ... View more

Your dates are dates, not strings looking like dates? ... View more

Do yourself a favour and don't overwrite datasets during code development, to avoid having to restore the dataset to its original state.   Could you please post some obs of the dataset your are processing? That may allow us to spot issues with the code more easily. ... View more

Hardly possible to understand what you want. Please explain "Deidetifying" / "decidentify", show the expected result and explain what you want to achieve. ... View more