macro variable during data step

[Brian3]

Hi,

the situation I have is a list of macro variables like follows;

var_2030

var_2031

var_2032

etc.

During the data step, which involves a loop, I need to use more than one:

data test;

x=2;

do i=1 to 3;

%let t=2030+i;

If x<&&var_&t then put 'success';

%end;

run;

Unfortunately, &&var_&t resolves as (var_2030) + i instead of (var_(2030+i)).

I've tried using %let t=%sysevalf(2030+i) but it then complains that there is a character variable instead of numeric. I also understand this may be related to macro values being evaluated before data steps are executed, so is this possible?

Thanks very much.

[Reeza]

From what you've posted here you should be using an array not a macro.

That being macro variables are not resolved in the same step they're created. THere are many ways to get around this, but given that you don't need macros my first suggestion is to switch to arrays instead.

Here's a tutorial on using Arrays in SAS
https://stats.idre.ucla.edu/sas/seminars/sas-arrays/

---

@Brian3 wrote:

Hi,

the situation I have is a list of macro variables like follows;

var_2030

var_2031

var_2032

etc.

During the data step, which involves a loop, I need to use more than one:

data test;

x=2;

do i=1 to 3;

%let t=2030+i;

If x<&&var_&t then put 'success';

%end;

run;

Unfortunately, &&var_&t resolves as (var_2030) + i instead of (var_(2030+i)).

I've tried using %let t=%sysevalf(2030+i) but it then complains that there is a character variable instead of numeric. I also understand this may be related to macro values being evaluated before data steps are executed, so is this possible?

Thanks very much.

---

[Brian3]

Thanks, for the task I'm doing I have to leave them as macro variables. Is there a way to do that?

[Kurt_Bremser]

---

@Brian3 wrote:
Thanks, for the task I'm doing I have to leave them as macro variables. Is there a way to do that?

---

No. Macro variables are resolved during data step compilation, so there's no way you can dynamically address them depending on values present while the data step executes. This is a very typical beginner's misconception when they do not yet understand the nature of the macro language as a code generator, not a data processing tool.

Please give us data and the expected result.

[Reeza]

---

@Brian3 wrote:
Thanks, for the task I'm doing I have to leave them as macro variables. Is there a way to do that?

---

Then you need to provide a more representative sample of your issue. 

[Kurt_Bremser]

Your code can't run anyway because of syntax errors. You have a do without an end, and a % end without a %do.

And if it was just a typo and it is a %do, that is not possible in open code, and I see no macro definition around your code.

From what I can guess, I take it that you have a bad dataset design (wide instead of long/tall), forcing you to write complicated code for something that is most probably an extremely simple task in SAS, needing no macro code at all.

So you first need to make your data intelligent (Maxim 33) by transposing; for further help, please provide example data in usable form (data step with datalines), and what you want to get as a result out of it.

[Brian3]

There should not be a %. It should be end.

[Brian3]

it is more the principle I am trying to do rather than specific data. Basically how to work with macro variables which can change value during a data step. Surely SAS has some ability to let me do that. Maybe I can use arrays, but not sure how it works with macro variables.

[Reeza]

What makes you think you need macro variables at all.

[Brian3]

its just because they are already macro variables. If there's a quick way to convert them to data values, that would be ok.

[Reeza]

Use the macro variables to define your array definition then, which is dynamic and then you can use the array without any issues. First develop working code - non macro. If you show us what that looks like we can help you convert it to a macro. Right now we have an xy problem.

[Brian3]

Hmm I've never used arrays before. Ok so

data test;

x=2;

%let var1=4;

%let var2=1;

%let var3=2;

array &var[3];

do i=1 to 2;

If x>&var[i] then put 'success';

end;

run;

[Tom]

No need for an array there as you are just dealing with numeric constants and not variables.

%let var1=4;
%let var2=1;
%let var3=2;

data test;
  x=2;
  if x > min(&var1, &var2, &var3) then put 'success';
run;

Much easier if you just put the list of values into a single macro variable instead.

%let list=4, 1, 2;

data test;
  x=2;
  if x > min(&list) then put 'success';
run;

Note: Make sure to move your %LET statements BEFORE the DATA statement. That is when they will execute so placing them there in the code will be less confusing to you.

[Kurt_Bremser]

If you want to create an array out of macro variables, see this:

%let var1=4;
%let var2=1;
%let var3=2;

data test;
x = 2;
array var[3] (&var1, &var2, &var3);
do i = 1 to 3;
  If x > var[i] then put i ', success';
end;
run;

• the %let statements are now where they belong to reflect the point in time where they are resolved
• the macro variables are used at a static point resolved during data step compilation
• the array is defined correctly (without the erroneous macro variable reference)

[Reeza · Accepted solution]

---

@Brian3 wrote:

Hmm I've never used arrays before. Ok so

data test;

x=2;

%let var1=4;

%let var2=1;

%let var3=2;

array &var[3];

do i=1 to 2;

If x>&var[i] then put 'success';

end;

run;

---

The X-Y Problem, as it is sometimes called, is a mental block which leads to enormous amounts of wasted time and energy, both on the part of people asking for help, and on the part of those providing help. It often goes something like this:

• User wants to do X.
• User doesn't know how to do X, but thinks they can fumble their way to a solution if they can just manage to do Y.
• User doesn't know how to do Y either.
• User asks for help with Y.
• Others try to help user with Y, but are confused because Y seems like a strange problem to want to solve.
• After much interaction and wasted time, it finally becomes clear that the user really wants help with X, and that Y wasn't even a suitable solution for X.

The problem occurs when people get their train of thought stuck on one approach and become unable to take a step back. Remaining open to having a new look at the bigger picture, these people might find their way back to X and continue searching for alternative solutions.

https://mywiki.wooledge.org/XyProblem