It would help if you read my instructions for providing data, you have now provided data in two different ways that I said NOT to do, and you haven't followed the examples and instructions in my post. ... View more

Thanks, but I can't use data in that format. I gave specific instructions about how to provide data, and also said not to provide data in Excel files and not to provide data as copy/paste from Excel. ... View more

It's always better to provide sample data (it could be fake data, as long as it illustrates the problem), as working SAS data step code (examples and instructions) and not as Excel files and not as copy and paste from Excel or elsewhere.   As an alternative, illustrate the problem you are having with packaged data sets that everyone has such as SASHELP.CARS or SASHELP.CLASS or any other data set in library SASHELP.   We consider this to be mandatory in your future posts.   Here is my solution using SASHELP.CARS   proc report data = sashelp.cars split='~'; column (" " make ) type,origin,msrp dummy; define make / 'MAKE' group order = internal; define type / '' across order = internal; define origin/ '' across order = internal; define msrp/mean 'MSRP'; define dummy/ noprint; run;   ... View more

It sounds to me like you could re-phrase the requirements to keep strings which contain digits or numbers, but not anything else. And so this is a job for the FINDC function.   if findc(test,,'kad') then delete; ... View more

Hello, @hmlong25  Any time code doesn't work right, we need to see the log, as @Amir said (whether its macro code or not). Use the macro debuggin command options mprint;  before your run the macro. ... View more

Question 1   'FEB10'd is not a valid SAS data literal. SAS does not understand this. Date literals MUST (as in, not optional, this is mandatory) be a two digit day followed by a 3 letter month, followed by a 2 or 4 digit year (and good practice means you will use a 4 digit year). No exceptions.   So you can write '01FEB2010'd, SAS would understand this. But even so, I cannot yet write code for you, because I don't know what the variable MONTH contains. If it contains numeric SAS date values, representing the number of days since 01JAN1960, formatted in any way you want, then the code is:   If month >= '01FEB2010'd then x1=1;   However, if the variable MONTH is character, or a numeric number which is not a valid SAS date value, such as 201002 (which could be interpreted by humans as February 2010 but would not be interpreted by SAS as February 2010), then I need specifics about what variable MONTH contains.   Question 2: Same problem, same answer, SAS does not understand 'MAR2010'd.   ... View more

This is a job for PROC SUMMARY. No need to write your own DATA step code with a mathematical formula.   proc summary data=example nway; class id; var ais; output out=want uss=sum_of_squares; run;   ... View more

Suppose you have an iterative procedure, it keeps going until some threshold is crossed.   do while (quantity >= threshold);   or   do until (quantity <= threshold);   ... View more

The variable named LENGHT is probably a mis-spelling of LENGTH, and this improperly spelled variable appears in your code. This mis-spelled variable has never been assigned a value so you see it is always missing.   NORMAL(0)*2 is a random number generator, multiplied by 2.   LENGTH always has a value of 100 because nowhere in the code does the value get changed, probably because of the mis-spelling that I mentioned. ... View more

When I look at the viewtable for my file, I see these exact characters: 06/08/2011 for dob and these exact characters: 18DEC23:18:48:18 for interviewdate.    What you see is either numeric variables formatted to 06/08/2011 or to 18DEC23:18:48:18; or these are character variables. We need to know which these are.   When I look at the viewtable after running the syntax you posted I see these exact characters: 18786 for dob and 2018544498.   These are numeric and unformatted. They are the dates and times that I asked for, unformatted. If you want to see what date and time they represent, you apply formats to them, you will see they are the exact dates and times that I asked for.   Maybe relevant. The dataset is exported from Redcap as a csv file and then read into excel and saved as an excel data file (xlsx) and that xlsx file is imported into sas through proc import (dbms=xlsx).   I don't think this is relevant. The relevant information we need is what is in your SAS data set. Please, we've gone long enough, please provide the actual SAS data set, using these instructions. Do not provide data in any other format — Excel is not acceptable, copy and paste from Excel is not acceptable. ... View more

This works for me, the YRDIF function gives a number around 12.5.   data a; interviewdt=dhms('18dec2023'd,18,48,18); dob='08JUN2011'd; teen_age=yrdif(dob,datepart(interviewdt)); run; ... View more

A data step is an implied loop, it will execute (as in a loop) for all records in the data set in the SET statement. So its not clear why you want a DO loop here, could you please explain? ... View more