SIPmath Standard reversed.png

SIPmath™ Standards

SIPmath Standard.png
 

Conveying Uncertainty as Data That Obey Both the Laws of Arithmetic and the Laws of Probability

330px-Wikipedia-logo-v2.svg.png

From Wikipedia:

The discipline of probability management represents uncertainties as auditable data that obey both the laws of arithmetic and the laws of probability. This is accomplished through arrays of simulated trials called Stochastic Information Packets (SIPs). A set of coherent SIPs which preserve statistical relationships is a Stochastic Library Unit with Relationships Preserved (SLURP).

SIPs and SLURPs allow your organization to network its simulations.

The SIPmath 2.0 Standard stores the data directly.

SIPmath 3.0 extends the SIPmath 2.0 by supporting virtual SIPs that incorporate the HDR portable pseudo random number generator and the Metalog distribution.

The 4 As of SIPmath

Coherent SIPs of either type are Actionable, Arithmetical, Auditable, and Agnostic.

Actionable

Actionable.png

SIPs are actionable in that they may be used as direct input to computer simulations. Like electricity, they may be generated by experts in one location for users in any other location.

To use the analogy of making toast in a toaster, knowing you need 900 W of 60-cycle AC current at 120 volts to make toast is NOT actionable. Plugging your toaster into an outlet IS actionable.

Arithmetical

Arithmetical.png

If the statistical relationships between SIPs are preserved, that is, they form a Stochastic Library Unit with Relationships Preserved (SLURP) and they have the same dimension (number of trials), then they may be added, subtracted, multiplied, or divided element by element to perform the arithmetic of uncertainty. This is what allows diverse simulations running on various platforms to be networked through their inputs and outputs to create a stochastic rollup.

Auditable

Auditable.png

Since SIPs consist of numerical data, they may be audited at any level of a simulation or network of simulations, and contain metadata that includes provenance or pedigree.

Agnostic

Agnostic.png

As data arrays, SIPs may be implemented in most software platforms. Current file types supported are Excel, CSV, XML, and JSON.