About pkopersk

novinosrin · ‎02-06-2020

Hi @pkopersk Your approach is correct, however you need an extra pass of the by group to assign the computed cumulative count to each record of the by group. So, Basically I have tweaked your code to read the BY group again by using a technique called interleave aka sorted append and conditionally compute the cumulative count and assign appropriately. data have; input VAR1 $ VAR2;* COUNT_BY_GROUP; cards; A 10 2 A 20 2 B 243 1 C 4 3 C 21 3 C 4 3 ; data want; set have(in=a) have(in=b); by VAR1; retain COUNT_BY_GROUP; if a then do; if first.VAR1 then COUNT_BY_GROUP = 0; COUNT_BY_GROUP +1; end; if b; run; The genius @mkeintz taught me this among many others he taught me personally over the last couple of years. Have fun learning!

FreelanceReinh · ‎09-19-2019

Hi @pkopersk, In addition to the issue with the macro calls resolving to invalid syntax (like "PROC step in a DATA step"), as pointed out by Kurt Bremser, your LENGTH statement should use the DATA step variable MTH, not the macro variable M in order to avoid truncation of the month names (longer than "janvier"). Personally, I would avoid national language characters in SAS names (my SAS version didn't even tolerate the accented characters in dataset names). You can use the BASECHAR function for an easy conversion: data _null_; length mth $9; do yrs = 2018 to 2019; do mth = 'janvier', 'février', 'mars', 'avril', 'mai', 'juin', 'juillet', 'août', 'septembre', 'octobre', 'novembre','décembre'; call execute(catx(',','%nrstr(%import_excel(C:\Users\a0782369\Desktop\testluk',yrs,basechar(mth),'RawData))')); end; end; run;

novinosrin · ‎09-18-2019

Without a SORT data have; input data_set variable_name$ DATA_A$ DATA_B$ DATA_C$ DATA_D$ ; cards; 1 A num . . . 1 B num . . . 1 C char . . . 1 D num . . . 2 A . num . . 2 B . num . . 2 C . char . . 2 D . num . . 3 A . . num . 3 B . . num . 3 C . . char . 3 D . . num . 4 A . . . num 4 B . . . num 4 C . . . char 4 D . . . num ;;;;; run; data _null_; if _n_=1 then do; dcl hash H (dataset:'have(obs=0)',ordered: "A") ; h.definekey ("variable_name") ; h.definedata ("variable_name","DATA_A", "DATA_B", "DATA_C","DATA_D") ; h.definedone () ; end; set have end=l; array t(*) DATA_A--DATA_D; array u(4)$ _temporary_; do _n_=1 to dim(t); u(_n_)=t(_n_); end; _n_=whichc(coalescec(of u(*)),of u(*)); rc=h.find(); t(_n_)=u(_n_); h.replace(); if l; h.output(dataset:'want'); run;

ballardw · ‎07-25-2019

@pkopersk wrote: How would I do that? I tried this: data WANT; infile "path\file.csv" delimiter= ';'; run; But the imported dataset is empty Use proc import with a csv, use a: guessingrows=max; statement. The log will have data step code generated by proc import to read the data. Copy the code from log and paste into the editor. Clean up by removing line numbers if such appear. Change the INFORMAT statements for the numeric variables, likely something like INFORMAT VARIABLE 32.; to a character informat like : INFORMAT VARIABLE $15. ; use a number large enough to read the longest expected sequence of digits. You can remove any of the FORMAT statements for these or other variables with a $w. informat. You should leave anything related to dates, times or date times alone as you will find in the long run that the SAS date values are much easier to manipulate or examine than strings. These would likely have informats that look like MMDDYY10. DATE9. ANYDTDTE40. TIME8. or similar depending on the layout of the variable. SAVE the code, especially if you will have other similar files read. Then you only need to change the infile statement and the output data set name.

Kurt_Bremser · ‎06-25-2019

Use update in a data step: data want1; update have1 have2; by id; run;

Patrick · ‎02-28-2019

Your code works for me. What error do you get? %let sysd= %sysfunc(day("&sysdate"d)); %let sysmonth= %sysfunc(month("&sysdate"d)); %let sysyear= %sysfunc(year("&sysdate"d)); %put &sysd &sysmonth &sysyear; data _null_; startdate=today()-55; enddate=mdy(&sysmonth,&sysd,&sysyear); vehi_age=datdif(startdate,enddate,'act/act'); put vehi_age; stop; run;

SuryaKiran · ‎01-24-2019

Hello, When you have to filter dates in SAS then you need to pass a date value, i.e. the value you have passed need to be understood by SAS. Leave the date value as numeric without formatting it, if you want to use a formatted value then convert that numeric value into your formatted vaue. %let year=2018; %let month=11; %let date=%sysfunc(mdy(&month,01,&year)); %put &=date; %let date_formated=%sysfunc(mdy(&month,01,&year),mmddyy10.); %put &=date_formated;

Astounding · ‎12-18-2018

Even once you resolve that error, there is an additional reason the code will not work. You cannot create &A within a DATA step, and then use it within the same DATA step. Basically, SAS has to determine what your statements are all about, before it has created &A. You would be better off keeping all your code in macro language: %let annee=2018; %let mois=11; %let a = %substr(&annee, 3, 2); %let suiviassinc = dsi_asti.suiviassincnew&a&mois; Even if you must use a DATA step for some other reason, you can emulate the %LET statement with: data _null_; call symputx('suiviassinc', "dsi_asti.suiviassincnew&a&mois"); run;

mkeintz · ‎11-27-2018

I often use date literal expressions as macrovars, but don't include the quotes (whether double or single) in the value assigned to the macrovar. Instead %let dfinactg=31jul2018; Then you can data suivi; set suivi; attrib rundate format=ddmmyy10.; rundate="&dfinactg"d; runyear=%sysfunc(year("&sysdate9"d)); run; In other words, construct the user-defined macrovar dfinactg the same as the system defines variable sydate9.

FreelanceReinh · ‎10-11-2018

In Windows 7: Click Start button, then "Standard Programs" (this can also be accessed via Control Panel), then "Associate a file type or protocol with a program". In the long list of filename extensions scroll down and select .sas7bdat, click "Change program" button and select SAS.UniViewer (if this is not listed, click Browse and navigate to the SAS.UniViewer.exe, which may reside in C:\Program Files\SASHome\SASUniversalViewer\1.4 [or higher version number] or wherever you installed it). Confirm with OK. While you're at it, you may want to associate file type .sas with SAS.UniViewer as well. I find this quite convenient.

Tom · ‎07-06-2018

Use functions. You don't really need a macro since it is just a one liner. But if you do make a macro then make it more useful by using parameters and creating a "function" style macro. %let annee=2018; %let mois=02; %macro bs(year,month); %sysfunc(intnx(month,%sysfunc(mdy(&month,1,&year)),0,e),date9) %mend bs ; %let dobs=%bs(&annee,&mois); Results: 589 %let dobs=%bs(&annee,&mois); 590 %put &=annee &=mois &=dobs ; ANNEE=2018 MOIS=02 DOBS=28FEB2018

Online Status	Offline
Date Last Visited	‎06-05-2020 06:45 AM

Create a count variable by group and output total next to each element...

Macro doesn't update variables after argument parsing in do loop

Aligning a dataset

Re: PROC IMPORT dbms= EXCEL force all imported variables to character

Re: PROC IMPORT dbms= EXCEL force all imported variables to character

PROC IMPORT dbms= EXCEL force all imported variables to character

Conditionnal merge of 2 datasets with replacement

Convert macro variable to numeric

Format of macro variable

Dynamically adapting the value of a macro variable using call symputx

Re: Macro doesn't update variables after argument parsing in do loop

Re: Aligning a dataset

Close all VIEWTABLES

Re: Create a count variable by group and output total next to each ele...

Re: Macro doesn't update variables after argument parsing in do loop

Re: Aligning a dataset

Re: PROC IMPORT dbms= EXCEL force all imported variables to character

Re: Conditionnal merge of 2 datasets with replacement

Re: Convert macro variable to numeric

Re: Format of macro variable

Re: Dynamically adapting the value of a macro variable using call symp...

Re: Put a single apostrophe in front of macro variable value

Re: Close all VIEWTABLES

Re: How to change the value stored in the macro variable change dynami...