Re: Statistical Analysis - ILERPG (V5R4)

Lots of math functions are available to use in ILE RPG, they are in QC2LE, but you don't have to bind to that anymore, it's done for us.

These are C++ functions, here's the link to an old post about it - http://as400pro.com/tipView.php?cat=RPG&key=251

There is also a set of CEE math APIs - see more at this link - https://www.mcpressonline.com/programming/apis/using-apis-to-extend-rpg-the-cee-apis

Get yourself a few statistical formulas, use these functions, if necessary - including trig functions and more - there you go!

Regards
Vern

On 4/2/2021 9:51 PM, Larry "DrFranken" Bolhuis wrote:

Well I remember my statistical analysis class in College. We were allowed any language. The school had Lisp, Assembler, BASIC, COBOL, FORTRAN. Assignment just had to be turned in before the start of class the day it was due. Assignment 1 a friend (today an IBM i Sr Systems guy) wrote his in COBOL. It was huge. Professor picked it up when he dropped it on the table. "What is THIS?" HE thumbed through the thing, it was at least 5 time bigger then the next biggest. Jim said: "It's COBOL."  Professor asked "Does it work?" Jim says: "Sure does. The output is at the end."  Prof slam dunked the thing in the trash. "Jim gets a 10 on this, but from now on NO COBOL!" At that point jim tapped me on the shoulder and whispered: "Glad he did that, i was never doing another assignment in COBOL!"

The best language at least then was FORTRAN for sure. BASIC wasn't horrible. Thinking about doing that in RPG II makes my head split. But in ILE RPG free form it seems like that would work just fine.

But then Susan took away my RPG card so I'd have to do it in CL today. That would be a horrible choice!
    - DrF

On 4/2/2021 10:24 PM, Chris Wager wrote:

Good Morning/Evening,

I know this question is a bit of a Hail Mary, but....

What do the seasoned RPG programmers amounts you think to the concept of
using RPG or specifically ILERPG to carry out a statistical analysis of a
given dataset.

For example to use data about five previous events to then generate what is
statistically probable to occur at event six.

Is this something that would be achievable with RPG or would I be
attempting to use a sledge hammer for a job requiring a tap hammer?

I mean its only numbers at the nuts and bolts level, so i am thinking it
should be feasible.

Open to all thoughts on this one.

Thanks
Chris