"If someone can code well in R, python or other similar programs, I think they could plug themselves in very quickly."   And if someone can program at a high level in SAS, they could plug in very quickly as well.   Experience with load forecasting seems to me to be the major criteria, not the programming language this person will use.   I think it is a mistake to phrase this as a question of which software to use. If you need an advanced/high level programmer, then you need an advanced high level programmer, regardless of which software that person uses. ... View more

@DocMartin  As you can see, there are a number of people not sure what to do here.   One of my concerns is that two examples of this bizarre layout of data may not be enough to cover all possible examples. And so the code that we provide based upon these two examples may fail on other data in this bizarre layout. Never a good approach.   This is why I ask for "logic" rather than "examples". Logic is a set of rules that will allow us to write code that could work for ALL possible input data. Logic > examples. ... View more

Repeating:  Similarly, I would like a better explanation of the logic of how anyone could start with your first table and wind up with the second table.   Also, the dollar signs have nothing to do with my objection about the layout of the 2nd table. LOS and Dollars do not belong in the same column. ... View more

I can read your words, but I don't see which of the letters in the table is hospital ID, further explanation is needed. Similarly, I would like a better explanation of the logic of how anyone could start with your first table and wind up with the second table.   In addition, in your second table, you have some values which are dollars in the same column as LOS. This is always something you should avoid, these should be in different columns for SAS to work most efficiently. ... View more

So, when I join tables A and test1, I do not get any data back.  There could be a dozen reasonable explanations for this. Please show us the code you are using. Please show us (a portion of) the data in table A and show us (a portion of) the data in table B. ... View more

Two simple steps   Find date of loan and find those offers within 7 days of loan (much simpler data step than those shown by others) Find max of offers within last 7 days (PROC SUMMARY); /* step 1 */ data intermediate; merge have(where=(loansamnt=.)) have(where=(loansamnt>.) drop=offeramnt rename=(date=loandate)); by custid; delta_date=loandate-date; run; /* step 2 */ proc summary data=intermediate nway; where delta_date between 0 and 7 or missing(delta_date); class custid; var offeramnt; output out=max_offer max=max_offer; run; ... View more

A change to your thought processes is not a solution.   Please be kind enough to help us understand and clearly state the solution to your programming issue. ... View more

@Jeff_DOC wrote: Thank you to everyone who took the time to reply to me. I accepted Cynthia's solution because that was the response I was reading when what Paige and Cynthia were saying finally got through to me.    I so wish I could split solutions into percentages. @Jeff_DOC  This is frustrating. We have tried to help you, but we want you to help us as well. Please clearly state the solution. It's clear that Cynthia and I did not present a solution; we both tried to debug this situation, neither of us offered a solution. ... View more

@TimurShangareev wrote: Is there any method, where I can extract dataset name from expressions like: <work.ds(...where=(...))> <ds(... where=(...))> or somehow check that this expression is valid in macro calls (wthout PROC or DATA statements)? Let me make sure I am understanding. You have some text, like   SASHELP.CLASS(where=(sex='F'))   and you want to determine the data set name (which is SASHELP.CLASS). Is that what you want?   Where do you want to do this? (In a SAS data step? Somewhere else?) What is the source of the input text string? ... View more

@colettb11  I repeat what @Tom said, use an array. ... View more

@Jeff_DOC  I requested the unformatted data. Thank you. I also requested the ENTIRE log for this data step (not just the errors or warnings, but every line in the log for this data step), which I still need. ... View more

Can we see the unformatted values, as I suggested, by using the PROC PRINT code I provided?   Problems like this with dates not doing what is expected are sometimes difficult to debug. This is why I am asking for specific information, we need you to provide the specific information I asked for, and not the information you decide is good enough.   Also, I cannot imagine where Division by 0 comes in here unless there's more to the code that you are not showing us. Can you show us the ENTIRE log for this data set (not just the warnings or errors) as I requested?   I do not get missings.   data abc; input (dob date_01) (yymmdd10.); age=int(yrdif(dob,date_01,'AGE')); format dob date_01 mmddyy10.; datalines; 1959-12-31 2024-01-11 ; proc print data=abc; var dob date_01 age; format dob date_01; run;   ... View more

I don't see a problem in this very small test:   data abc; input (dob date_01) (yymmdd10.); age=int(yrdif(dob,date_01,'AGE')); datalines; 1955-01-01 2025-08-15 ;   Format of the variables is irrelevant. Negative dates are irrelevant. If the dates are truly dates, then YEARCUTOFF= should be irrelevant.   Please provide the full data step or full SQL code, not a fraction of a line of code as in your original message. Also, if there are WARNINGS or ERRORs in the log, please show us the log for this data step or SQL (and not just the lines with the WARNINGs or ERRORs, we need to see the full log for this data step or SQL.)   Please provide a small sample of data that includes dates before and after 01JAN1960. Best would be if the dates are UNformatted (yes UNFORMATTED). Please provide the data as DATA step code, according to these examples and instructions as I have done above;  or I think this short program will work as well in this situation on a small subset of the data with dates before and after 01JAN1960. Do not provide the data via COPY/PASTE from Excel, do not provide the data as an Excel file.   proc print data=abc; /* Use your real data set name here */ var dob date_01 age; format dob date_01; /* Unformatted */ run;   ... View more

The answer I would give in the interview is the same answer I gave you. SQL is not the right tool for this. ... View more

@Hakob wrote: How do you find the 3rd highest salary using a self-join? Maxim 14 — Use the Right Tool. SQL is a poor choice. PROC RANK, or PROC SORT, will be easier to program and probably run faster. ... View more