About art297

art297 · ‎07-20-2018

@ChrisHemedinger: I'm a bit late to the discussion but, since the goal was to produce code that required the fewest keystrokes, I'm surprised that no one took advantage of SAS's automatic spelling correction that goes on when code is submitted. I liked one of @Tom's entries, thus shortened his code by three keystrokes: dat have; inpu k $12.; card; 000112010302 0 1 2 3 4 5 6 7 8 9 0123456987 . ; dat w2; se have; h=10-findc('9876543210 ',k); ru; Art, CEO, AnalystFinder.com

art297 · ‎07-17-2018

If your two files are already sorted by x I would use the following method: data one; input x $ y $; datalines ; 1 2 1 3 2 4 2 5 2 9 ; data two; input x $ z $; cards; 1 4 1 5 1 6 2 3 2 7 ; data one_a; set one; by x; if first.x then seq=1; else seq+1; run; data two_a; set two; by x; if first.x then seq=1; else seq+1; run; data want (drop=seq); merge one_a two_a; by x seq; run; @janus58: the placement of the call missing statement is critical. If it is the first assignment of those variables, it will make the fields numeric. That is why @Ksharp's code wouldn't produce a mixed case error, as the call missing will assign them as character (as they've already been assigned that way). However, in the present case, it isn't needed. Since it is a many to many merge, that is the problem that has to be overcome. I did that by assigning seq numbers and including that variable in the by statement. Art, CEO, AnalystFinder.com

art297 · ‎07-14-2018

I don't meet your criteria for being a top poster, but I think I can help answer your question above what has already been said. Of course, I can't speak for everyone, thus others will have to chime in with their agreement, disagreement, or additional rationale. I started using on-line forums because I needed answers that I couldn't find elsewhere. That quickly led to a paying-it-forward rationale because I felt that others would be more likely to answer my questions if they saw that I answered more questions than I asked. Then I discovered that answering others' questions was, in itself, one of the best ways to learn SAS. An unexpected motivating factor came in when, as a result of my participation, I was asked to co-author a number of papers and, when attending conferences, many people knew and welcomed me when I attended a meeting or event. Additionally, I think that the reputation I got from participating on on-line forums, has led to many expense paid trips to SAS Global Forum and many user group meetings. Participating in on-line forums is addictive. I intentionally don't meet your criteria as, after a number of years, I realized that my trying to beat others in answering posts was, in fact, keeping others from reaping the same benefits I've received. As such, I try to only answer questions these days if I notice that no one else has provided an optimal answer/solution. Of course, the addictiveness aspect sometimes conflicts with that rationale. While I have a Ph.D., have been using SAS for more than 45 years, and am now retired, I still consider myself a learner. SAS is always expanding and adding new functionality, so there is always something new to learn. Art, CEO, AnalystFinder.com

art297 · ‎07-12-2018

The following should work: data raw; informat text $80.; input text &; cards; abc(123) (123)abc(pqr) ; data want; set raw; want = prxchange('s/$([^$]+)\)//i', -1, text); run; Art, CEO, AnalystFinder.com

art297 · ‎07-11-2018

Your code (once modified to meet what proc sql expects) appears to work correctly. I ran: Proc SQL; create table work.covers_before_total as Select a.*, Sum(Case When a.Price_start_Date <= a.NEW_Price_End_Date and a.Price_start_Date <= b.date and a.NEW_Price_End_Date >= b.date then b.Covers Else 0 END) as covers from work.covers_before a,work.covers_measurement_model b group by a.id ; quit; Can't help further unless/until you provide some example data for the two datasets, and the result you expect. Art, CEO, AnalystFinder.com

art297 · ‎07-04-2018

Yes! The macro was designed to run on base SAS using a Windows operating system. Art, CEO, AnalystFinder.com

art297 · ‎06-30-2018

@novinosrin: I posted the IW workbook sometime ago (see: https://communities.sas.com/t5/Community-Matters/Can-we-have-best-excerpts-of-SAS-L-archives-in-our-library-here/td-p/463586 ). In fact, it's attached to my response to your post. Art, CEO, AnalystFinder.com

art297 · ‎06-30-2018

see: https://communities.sas.com/t5/Base-SAS-Programming/How-to-import-data-from-excel-xlsx-sheet-with-headers-and/td-p/246864 Art, CEO, AnalystFinder.com

art297 · ‎06-27-2018

If your data are already in the order shown in your example, then you could use something like: data need; set test; by modelo name_stat concept notsorted; if first.concept then order=1; else order+1; run; PROC SORT DATA=need; BY MODELO NAME_STAT order; RUN; PROC TRANSPOSE DATA=WORK.need out=want(DROP=_NAME_ order); BY MODELO NAME_STAT order; VAR VALUE; ID CONCEPT; RUN; Art, CEO, AnalystFinder.com

art297 · ‎06-26-2018

@MelodieRush: It was Variable Selection: SAS Enterprise Guide & SAS Enterprise Miner. Thanks for letting me know! Excellent seminar! Art, CEO, AnalystFinder.com

art297 · ‎06-26-2018

@MelodieRush: Just watched a different seminar and was looking for the PDF, but can't find it on the ask-the-expert page. The title was something to the effect of Data Reduction using EG and EM. Has that been posted somewhere? Art, CEO, AnalystFinder.com

art297 · ‎06-25-2018

@ChrisHemedinger: If you want to write an article about a "really" undocumented, yet even more (methinks) useful procedure, do one on proc spell. It has the same listing problem for SAS UE, but easy to get around: options caps; filename bigmsg temp; ods listing file=bigmsg; filename temp temp; data _null_; file temp; informat sentence $100.; input sentence &; put sentence; cards; Let's see if sas spell procdure can be used to verify whether tha seperate words in this, uhm, flie are, uhm, valid against a stantard internal dictionary and let’s see how versatile it is ; proc spell in=temp nomaster; run; proc spell in=temp verify; run; proc spell in=temp suggest; run; ods listing close; data _null_; infile bigmsg; input; put _infile_; run; Art, CEO, AnalystFinder.com

art297 · ‎06-25-2018

@LaurenceLetarte: The latest UE update appears to have corrected the problem regarding newer versions of Chrome. Art, CEO, AnalystFinder.com

art297 · ‎06-22-2018

You can't run a proc from within a datastep, but you can use call execute. And I don't understand what you're trying to merge, as all of your data records will either have number eq 1 or 2. I think the following is similar to what you want: /* Create test files */ proc sort data=sashelp.class out=class; by name; run; data class; set class; if sex eq 'M' then number=2; else number=3; run; data jmp2 (keep=name height); set class (where=(sex eq 'M')); run; PROC EXPORT DATA= WORK.jmp2 OUTFILE= "/folders/myfolders/jmp2.jmp" DBMS=JMP REPLACE; RUN; data jmp3 (keep=name weight); set class (where=(sex eq 'F')); run; PROC EXPORT DATA= WORK.jmp3 OUTFILE= "/folders/myfolders/jmp3.jmp" DBMS=JMP REPLACE; RUN; /* Create table1 for number 2 */ data table1; set class(keep=name sex number where=(sex eq 'M')); run; /* Run code for number eq 2 */ data _null_; set table1(obs=1); call execute('proc import out = work.temp dbms=JMP replace datafile='); if number eq 2 then do; /* call execute('C:\JMP2.jmp;run;'); */ call execute("'/folders/myfolders/JMP2.jmp';run;"); end; else if number eq 3 then do; /* call execute('C:\JMP3.jmp;run;'); */ call execute("'/folders/myfolders/JMP3.jmp';run;"); end; run; data want; merge table1 temp; by name; run; /* Create table1 for number 3 */ data table1; set class(keep=name sex number where=(sex eq 'F')); run; /* Run code for number eq 3 */ data _null_; set table1(obs=1); call execute('proc import out = work.temp dbms=JMP replace datafile='); if number eq 2 then do; /* call execute('C:\JMP2.jmp;run;'); */ call execute("'/folders/myfolders/JMP2.jmp';run;"); end; else if number eq 3 then do; /* call execute('C:\JMP3.jmp;run;'); */ call execute("'/folders/myfolders/JMP3.jmp';run;"); end; run; data want; merge table1 temp; by name; run; Art, CEO, AnalystFinder.com

art297 · ‎06-20-2018

Read up on the modify statement (e.g., http://support.sas.com/documentation/cdl/en/lrdict/64316/HTML/default/viewer.htm#a000173361.htm ). Your answer will depend upon the combination of what your new file looks like and the updatemode you select. Without seeing your actual datasets it's difficult to say what will do exactly what you want. You could use merge, update or modify. And, for each, you use options to control what will happen. Art, CEO, AnalystFinder.com

Online Status	Offline
Date Last Visited	‎11-26-2024 04:45 PM