Terrific code from @ballardw, to create 'dummy variables': Bottom 10% of a certain variable, code dummy variable as 1. Otherwise, code the dummy variable as 0. But I've encountered a problem....   proc summary data=&etf..&etf._combined; var i_50401_Z i_50402_Z i_50403_Z i_50404_Z i_50405_Z i_50408_Z ; output out=&etf..&etf._combined_temp (drop= _:) p10= /autoname; run; Proc sql; create table &etf..&etf._combined_2 as select a.*, (a.i_50401_Z <= b.i_50401_Z_P10) as i_50401_Z_bottom10pct , (a.i_50402_Z <= b.i_50402_Z_P10) as i_50402_Z_bottom10pct , (a.i_50403_Z <= b.i_50403_Z_P10) as i_50403_Z_bottom10pct , (a.i_50404_Z <= b.i_50404_Z_P10) as i_50404_Z_bottom10pct , (a.i_50405_Z <= b.i_50405_Z_P10) as i_50405_Z_bottom10pct , (a.i_50408_Z <= b.i_50408_Z_P10) as i_50408_Z_bottom10pct from &etf..&etf._combined as a, &etf..&etf._combined_temp as b; quit;   The problem is, there is some missing data.  Blank cells.   And the above code is dummy coding these blank cells as 1.   So, I have to somehow tell the code above to only code for the lowest 10% of actual numerical values.  Ignore the blanks.   Any help greatly appreciated.   Nicholas Kormanik     ... View more

I accepted an answer a bit too hastily when I tried to ask a question earlier.  I wasn't clear enough with my question then.  And didn't understand the answers given.  So, need to try one more time....   @SASKiwi , @PaigeMiller , @Kurt_Bremser , and helpful community.   I've set up a complete document/program:   "Logistic, All.txt"   At the top of the document is:   %let etf = xyz ; Then all throughout the document is the following, as appropriate, everywhere "xyz" needs to appear:   &etf. SAS calls this trick creating and using a macro variable.  Pretty straight forward, easy enough, and incredibly useful.   Running the entire long program set in SAS works perfectly.  All is good with the entire amalgam of procedure steps.  Completion is as desired.   What's next is to act on a whole list, as "xyz" is only ONE of a hundred such.   The complete list, say, is: xyz abc def lmn qrs (etc.)   So, as of this second, to move forward, I would need to replicate the above-mentioned document 100x, manually changing only the "xyz" at the very top of each document, to the appropriate characters, and then running each of the 100 documents through SAS.  This can fairly efficiently be done in a single go (not 100 separate runs) -- using the %include approach, in an additionally created document:   "Logistic, All, Complete List, Using 'Include'.txt":   %include "c:\1\Logistic, All -- xyz.txt" ; %include "c:\1\Logistic, All -- abc.txt" ; %include "c:\1\Logistic, All -- def.txt" ; %include "c:\1\Logistic, All -- lmn.txt" ; %include "c:\1\Logistic, All -- qrs.txt" ; (etc.) Instead of going about it in the above manner, I'm pretty sure there is a macro approach.  Automatically get each of the list items over into the "xyz" position (at the very top), one at a time, through the entire list, running the entire procedure set at each iteration of each change.   I hope the task is clear enough this time around, and look forward to your responses.  Please try to answer as clear as day, as this all seems rather cloudy at present.  And hopefully others will see this post and learn something as well.  There may be more than one way of accomplishing the task.  Please share your thoughts.   Humorously, I don't even know what to call the above challenge, so as to conduct a successful Google search.   Thanks!   Nicholas Kormanik     ... View more

Just did some searching and reading on SAS log file directing.  And, phew!, seems to be a crazy topic, all across the board.   Simply put, I'd like to process several datasets, one at a time:   aaa.sas7bdat bbb.sas7bdat ccc.sas7bdat   Each dataset resides within its own folder:   c:\1\aaa\aaa.sas7bdat c:\1\bbb\bbb.sas7bdat c:\1\ccc\ccc.sas7bdat   (I'd like to process all the datasets using a macro.  But will inquire about how to do that tomorrow.)   For now, when processing aaa.sas7bdat, I would like to create a log -- log_aaa.txt -- within c:\1\aaa\.   Following completion of that process, I'd like to move on to bbb.sas7bdat.  And during the process create a new log file -- log_bbb.txt.  That file should reside in the respective c:\1\bbb\.   On and on....   You get the picture?   Any sage advice on how best to do this will be greatly appreciated.   Thanks!   Nicholas Kormanik     ... View more

The code below works just fine: libname SPY 'e:\00\spy' ; Log output: 1 2 libname SPY 'e:\00\spy' ; NOTE: Libref SPY was successfully assigned as follows: Engine: V9 Physical Name: e:\00\spy   The code below does not work: %let etf = spy ; libname &etf. 'e:\00\&etf.' ; Log output: 1 2 %let etf = spy ; 3 4 libname &etf. 'e:\00\&etf.' ; NOTE: Library SPY does not exist.   Help Mr. Wizzard.......     ... View more

Please see the following code:   ods output ParameterEstimates = sas_1.ParameterEstimates ;   In Proc Logistic, selection = stepwise, (and in perhaps other procedures), more than one ParameterEstimates files (datasets) are generated.  (Discovered via ods trace on.)   With each subsequent creation, SAS overwrites the existing ParameterEstimates file, if one exists.  So, what one ends up with (apparently) is the 'final' version.  (Please correct if I am mistaken here.)   With the generation of graphs, SAS auto-numbers incrementing new graphs with appended numbers.   Can we do similarly with the ods output to datasets?   If you wanted to catch each and every dataset output generated, how would you achieve this?  (Say, for instance, to compare the first with the final...)   Thanks!   Nicholas Kormanik   ... View more