@ari2495 wrote: data AUC_Base; set _smd_pd0base_imp _smd_pd0base_poe _smd_pd0base_pri _smd_pd0base_smb; run; data AUC_Weighted; set _smd_pd_weighted_imp _smd_pd_weighted_poe _smd_pd_weighted_pri _smd_pd_weighted_smb; run; And these lines as well ... as the macro produces intermediate tables reflecting the arguments given to it.   ... View more

If (and only if) your dataset meets the demands on its structure you needed to adjust these lines: @ari2495 wrote: /*BASE*/ %AUC_test (input = Discr , pd_est = PD0Base , default_flag = flag_default_finale, Segmento= IMP ); %AUC_test (input = Discr , pd_est = PD0Base , default_flag = flag_default_finale, Segmento= POE ); %AUC_test (input = Discr , pd_est = PD0Base , default_flag = flag_default_finale, Segmento= PRI ); %AUC_test (input = Discr , pd_est = PD0Base , default_flag = flag_default_finale, Segmento= SMB ); /*WEIGHTED*/ %AUC_test (input = Discr , pd_est = PD_weighted , default_flag = flag_default_finale, Segmento= IMP ); %AUC_test (input = Discr , pd_est = PD_weighted , default_flag = flag_default_finale, Segmento= POE ); %AUC_test (input = Discr , pd_est = PD_weighted , default_flag = flag_default_finale, Segmento= PRI ); %AUC_test (input = Discr , pd_est = PD_weighted , default_flag = flag_default_finale, Segmento= SMB ); Where Discr needs to be your input table, PD0Base and PD_weighted columns with the pds and the Segmento= needs to be given a valid segment code (i.e. one in your table). ... View more

Lovely. Does that person know how to create a dataset to work with it? There needs to be a MOD_RATING_ALLA_DATA_RIF variable containing codes for "segments" (i.e. those values in "segmento" further in the code). ... View more

Hello! Hugh ... long answer. First off ... the code seems to be intended to run as is, i.e. the macro is defined _and_ used there. The adaption to your tables is done in the /* BASE */ and /* WEIGHTED */ sections of the code, where the Macro is called. Nevertheless, there are some assumptions on the structure of your data coded there. My question is, how much do you understand of the macro code? I.e. can you "read" macros? And then: Do the SAS statements within the macro make sense to you? That was a way to keep the answers to your question short. And of course: The person who provided you with the code should be able to tell you how to use it. --FJa ... View more

which one can run in a data step: data _null_; infile pipe "split --bytes=30M &filename. 2>&1"; input; put _infile_; run; Awesome! I didn't think of that one ... But you are absolutely right! ... View more

Does Kurt's reply solve your Problem? ... View more

Again, let me draw your attention to the OS. In Unix it is a one line command: split --bytes=30M filename ... View more

Hello! I do not know about manipulating metadata to rename variables, but what about simply copying the vars and dropping the old? Would that be a fix for you? See the sketch below. Cheers data work.NewTestData; set work.TESTDATA; LovelyName = BadName; drop BadName; run;   ... View more

Hello! Do you need to do that in SAS? This is more conveniently done on OS level. --FJa ... View more

Hello! As it appears to me that you do not want to calculate the values yourself, why not having an array of values and looping over it ... checking for . ? Regards ... View more

Hello! A quick question: Does the code without the last to if clauses meet your expectations? i.e. does the code: data dil_merge; merge hw6.dil1 hw6.dil2_temp; by MEDRNO; if xwtgainlb > 60 then delete; if xwtgainlb < 60 then delete; run; return the correct results.  If yes, then you needed to check on xwtgainlb as @Kurt_Bremser has already suggested. --fja ... View more

Hello! If you do not write anything in case of MET value = .  to the output dataset, then you could use: if <condition> then delete; to continue with the next line. Hope I understood you correctly. --FJa ... View more

@Nietzsche wrote: I don't see no such code at the end of any of my logs Hello @Nietsche!   I see the following in my log: 70 data work.admitfee; 71 set cert.admit; 72 where actlevel='HIGH; 73 run; 74 75 proc print data=work.admitfee; 76 var id name actlevel fee; 77 run; 78 79 OPTIONS NONOTES NOSTIMER NOSOURCE NOSYNTAXCHECK; 80 ODS HTML CLOSE; 81 &GRAPHTERM; ;*';*";*/;RUN; NOTE: There were 0 observations read from the data set CERT.ADMIT. WHERE 0 /* an obviously FALSE WHERE clause */ ; NOTE: The data set WORK.ADMITFEE has 0 observations and 9 variables. Where line number 81 contains the added code as indicated by @yabwon and the Note contains that famous "warning". You should actually see something similar. Best Fja ... View more