OpenChainGraph Suite · ART-371 · Simulation & replay

Portfolio VaR: Monte Carlo (Integer PRNG)

Runs a one-factor correlated-asset Monte Carlo simulation of portfolio Value-at-Risk and Expected Shortfall. The entire path loop, including the pseudo-random draws (xoshiro256**) and every arithmetic step, runs on integers in fixed-point, never a floating-point accumulator, so a declared seed reproduces the identical result every time. Declares the OpenChainGraph estimated determinism class (SPEC.md §24.6): stochastic output, exempt from cross-surface byte-identity, still required to be finite and idempotent for a fixed seed.

Simulation & ReplayInteger PRNG · fixed-pointZero PIIW3C VC §13.11
🔒 All inputs are processed locally in your browser. No data is transmitted. Do not enter real personal data — use synthetic or anonymised inputs only.
Estimated, not bit-exact
This kernel declares the SPEC.md §24.6 estimated determinism class, not bit-exact. It is exempt from cross-surface byte-identity but is proven, in the shipped fixtures, to replay byte-identically for a fixed declared seed and to change output when the seed changes. It is not a claim of parametric or historical-simulation VaR accuracy: it is a Monte Carlo estimate over a simplified one-factor correlation model.
Determinism inputs
PRNG algorithm: xoshiro256** (splitmix64 seed expansion), integer-only. Every path draw and every arithmetic step in the simulation loop is fixed-point (scale 1e6), never a JS float accumulator. The declared seed, prng_algorithm, and draw_count are carried as ordinary fields in the receipt below.
Presets
Simulation Parameters
Result
Execution Hash & §4 Artifact
SHA-256 execution hash (JCS canonical, RFC 8785):