Thank you for all your answers.
I've studied COBOL but I've never seen nothing about intermediate results.
When I search in google "COBOL handles intermediate results" I come to some pages, one being IBM:
https://www.ibm.com/support/knowledgece ... inr02.html
The page is interesting, but it doesn't show the reason I should use intermediate results.
If I understood well, if we have variables A with PIC 9(05), B, C and D, all three with PIC 9(02) and I code:
COMPUTE A = (B + C) * D
In the above case (B + C) is an intermediate result and after is multiplied by D it gives the final result.
But I don't see what's the advantage in using the intermediate result.
I'm still a COBOL "student" and when I don't know something, I will admit it instead of pretending to know more than I actually do.
Thank you,
Roger