Changing Data Output from Columns to Rows

[novinosrin]

Nothing is too much. I love these questions Gonna go out for lunch to give farewell to my classmate who just graduated. Will respond after lunch(Chicago time).Hang in there. Cheers

[novinosrin]

@new_sas_user_4

The set statement reads one record into a memory area known as program data vector that holds the observation while datastep is executed
So when set have executes for the 1st time
A 10 100 1
is read into memory and is held for further processing i.e using other statements that
follows set statement
Now let's get into the loop.
The index variable n when 1,
references the 1st element of the array as j(n)<=>income
Inc_rain_snow=vname(j(n));
this resolves to Inc_rain_snow=vname(j(1))
and that is Inc_rain_snow=vname(income);
similarly
value=j(n);
resolves to value=j(1); and that is value=income;
the rest of the variable values remain undisturbed and the output statment writes the
contents in the memory to the output dataset. The same happens for 3 iterations as we loop through
the 3 elements in the array. So the output statement executes thrice within the loop writing the
value of area A thrice.

Once SAS has executed all the statements for the 1st records, the process repeats for the next record
until end of file as it goes back to the top of the datastep.

[new_sas_user_4]

Thank you so much @novinosrin for taking the time to explain this 🙂

I got it 🙂

[Tom]

Why do you need to change the order of columns?

The normal SAS "trick" to changing the order of the columns is to use a RETAIN statement, since it does not imply a type.  

data want;
   retain _name_ area col1 ;
   set have;
run;

Or you could use PROC SQL.

proc sql;
  create table want as 
  select 
 _name_
, area
, col1
  from have
  ;
quit;