1

I recently read something that talks about DIST distribution strings. It appears to be a way to take a long string of previously generated numbers and somehow compress them into a string that can then be used by others. I believe it's lossy. I'm having a great deal of trouble finding out anything other than some generalities about it. It may be proprietary, I don't know.

Does anyone know where I might be able to find something out about the algorithm used to create such a DIST string? Also if anyone knows anything about this, is this something that's being used anywhere and how useful does it appear to be?

Mitchell Kaplan
  • 111
  • 1

1 Answers1

1

Online, there doesn't seem to be much information, although it's supposedly described by its inventor, Sam Savage, in its book The Flaw of Averages.

Apparently, it's a lossy XML standard, according to this website. It looks like a proprietary standard managed by Vector Economics and Probability Management. If this standard is a key piece of intellectual property for Sam Savage, I doubt the standard is publicized anywhere.

It appears to be used in the financial services industry, but for scientific purposes, the lack of openness may be a big problem when it comes to checking numerical results.

Geoff Oxberry
  • 30,394
  • 9
  • 64
  • 127