About Kurt_Bremser

Kurt_Bremser · ‎10-28-2014

Since the format for the zip code and state code is given, I recommend to roll up the "address line 2" string from the back, starting with length()-x, where x is sufficent to place you at the start of zip/state. What's left is then the municipality name.

Kurt_Bremser · ‎10-28-2014

How about: proc sort data=have; by CustCode ProdCode descending ProdRunDateKey; run; data want; set have; by CustCode ProdCode; if first.ProdCode; run;

Kurt_Bremser · ‎10-28-2014

This works: data have; infile cards dlm=','; format No 3. CustomerID $6. Product $25. Category $3. Age $6. Income commax10. DSA $6. City $3. SECODE $2. SEAGE 3. Qualification 2. Pricing $10. ; informat income commax10.; * because of dot; input No CustomeriD Product Category Age Income DSA City SECODE SEAGE Qualification Pricing ; cards; 1,MUM001,Savings Account,NMC,21-30,388.662,ZARETH,MUM,Z9,59,4,Discount 2,MUM002,LIC,NMC,51-60,98.530,CRACKF,MUM,C2,37,2,Discount ; run; proc sql; create view want as select product, category,income, age, case when age eq 'gt 60' then 'Top' when age eq '51-60' then 'Medium' when age eq '41-50' then 'Below Medium' else 'Low' end as Aged from have; quit; proc print data=want noobs; run; and renders the following result: Product Category Income Age Aged Savings Account NMC 388.662 21-30 Low LIC NMC 98.530 51-60 Medium

Kurt_Bremser · ‎10-28-2014

Before going through all the hoops, just write a simple data step with datalines as wanted. Much easier to read and decipher in the future. "When in doubt, use brute force" - Ken Thompson, creator of UNIX

Kurt_Bremser · ‎10-28-2014

Does not work, at least in 9.2 (the puts from the log, using exactly your code) now=2014-10-28:07:16: now2=28OCT2014:07:16:55.128662 In which environment did you test it?

Kurt_Bremser · ‎10-27-2014

What is your ERROR message and could you _please_ supply test data?

Kurt_Bremser · ‎10-27-2014

The OP tried to use dim(regs(z)) which does not work because regs(z) is not an array, but a single variable

Kurt_Bremser · ‎10-27-2014

Since the picture format directive for seconds (%s) displays only 2 digits for seconds, and one cannot combine the directives with .9999 for digits after the comma, you may need to employ my previously posted method.

Kurt_Bremser · ‎10-27-2014

Yeah, but x(1) will not give you the array a, but a1. So the whole thing lends itself to the misconception of the OP.

Kurt_Bremser · ‎10-27-2014

You cannot have an arrray of arrays, only an array of vars. But you can have multidimensional arrays. Although I can see no sense in your overly complicated code. Just write a data step that generates your regions in geogr2;

Kurt_Bremser · ‎10-27-2014

This depends very much on the way you need to do the upgrade, and what your environment looks like - Is your server only for SAS (that means: no other apps can cause conflict (eg port numbers) with SAS) - Will you set up a completely new server/client environment? In that case you can do the install out of the box on the new server (and possibly clients), without altering things like install path (esp. the configuration in the SAS subdir) or port numbers. Manually selected port numbers can cause lots of grief - Do you want to migrate your metadata or will you create new metadata (eg if you found logical shortcomings in your 9.2 structure) SAS only server with a completely new, "green field" installation could probably be done in one or 2 days. Complex installations (multiple-use server, several concurrent SAS versions) can take months (I know from dire experience) until you've weeded out all the errors.

Kurt_Bremser · ‎10-24-2014

Quick, without testing: data intermed; set have; retain company; if field2 = "company" then company = field1; else output; run; proc sort data=intermed; by company field2; run; data want (keep=company category mar_14 jun_14); set intermed (rename=(field2=category)); by company category; if first.category then do; mar_14 = 0; jun_14 = 0; end; select (date); when ('Mar-14') mar_14 = field1; when ('Jun-14') jun_14 = field1; otherwise; end; if last.category then output; run; In the last data step, the possible dates may be covered by a suitable macro.

Kurt_Bremser · ‎10-24-2014

And drop the erroneous equal sign

Kurt_Bremser · ‎10-24-2014

newval = put(datepart(oldval),yymmddd10.)!!'-'!!put(timepart(oldval),time15.6); You may need to take care of blanks for hours<10 fixed "d" for dash/hyphen

Kurt_Bremser · ‎10-24-2014

And oh yeah, doing a defragmentation of your C: drive may help a little.

Online Status	Online
Date Last Visited	7 hours ago