1000 possible predictive variables; one target variable -- PROFIT.   Looking for a way to whittle down the possible predictive variables.   Ideally I'd like to end up with one single best.  Or a handful.   Last year someone suggested using HPSplit. The results obtained seemed inconclusive. Many, many parameters to guess at in that procedure, and I might have gotten most guesses wrong.   Lately I've been coming across lots of mention of XGBoost -- a new whiz kid on the algorithms block.   Wondering if you all would recommend using that?   Since XGBoost is not natively included in SAS 9.4, however, trying a built-in procedure would be preferred.   Any thoughts greatly appreciated.   Nicholas Kormanik   ... View more

I think this is a 'macro' question.  Have very little experience with that.  Seems a bit overwhelming.   Please see the following code:   ods html close; ods html5 path="c:\0_SAS_1" file="i_20201.htm"; proc freq data=sas_1.variable_values_separate_runs nlevels order=freq; where Indicator = "i_20201"; tables indicator * range / nocum nocol norow crosslist out=sas_1.i_20201; weight number_weighted_yes / zeros; run; ods html close; The code above focuses on a variable i_20201.   Suppose I have 100 different variables that I'd like to focus on:   i_00211 i_00235 i_00249 i_00250 i_00262 ... i_00275 i_00283 i_00290 i_00296   So, I have a list of these other variables.  Same program as at the top here.   I need to replace i_20201 with each subsequent variable, and run the program again.   My hunch is to use a 'macro variable' statement at start of program:   %LET xyz = i_20201;   And wherever "&xyz." appears in the program, i_20201 will appear replacing it.  (By the way, notice the little period following xyz?  Super important.  Spent hours on that one.)   Helpful, sort of.  But not the whole job.  I need to get each of the 100 different variables to be placed there, respectively, for subsequent runs.   Is there a relatively easy way of going about this?   Thoughts greatly appreciated.   Nicholas Kormanik               ... View more

See the following code:   proc import datafile="c:\2\(not sure yet).csv" out=sas_1.really_want_this_data dbms=csv replace; getnames=no; guessingrows=max; run; Is there any way of asking SAS to simply grab the .csv file in C:\2\, and use whatever name it has, since I don't yet know the name of that file (or the name changes all the time)?   It is the one and only file in C:\2\.   Thanks much!   Nicholas Kormanik       ... View more

I'm surprised to see all the non-answers in documentation and posts regarding a simple question:   Where to store your macros??   Lots of good macros out there, maybe hundreds, thousands.  Suppose we collect a fair number of them.  And, maybe even write a few... or one.  (so far I'm at zero)   Well, where best to keep all those macros?.  (Assume Windows computer.)   How about:  "D:\SAS Macros\"   And, follow-up.  How then do we access one of those macros we've saved?   Thoughts greatly appreciated.   Thanks, Nicholas Kormanik           ... View more

Imported .csv from Excel.  Several columns are as follows:   +25.49% +28.95% -31.39% +35.36% -24.07% -34.71%   Import has designated these as 'character.'  I need to change these to 'number,' and get rid of the +, -, %.   Suggestions greatly appreciated.   Thanks! Nicholas Kormanik   ... View more