stochastic-rs

A Rust library for stochastic process simulation, quantitative finance, statistics, copulas, distributions, and neural-network volatility models. Generic over the float type (f32 / f64), with SIMD acceleration on CPU and CUDA / Metal backends where they pay off, and first-class Python bindings via PyO3.

What's in here

The workspace ships nine sub-crates (the stochastic-rs umbrella re-exports everything):

stochastic-rs-core             simd_rng (foundation)
stochastic-rs-distributions    19 distributions, FloatExt + SimdFloatExt
stochastic-rs-stochastic       120+ processes, ProcessExt
stochastic-rs-copulas          bivariate + multivariate copulas
stochastic-rs-stats            estimators (MLE, Hurst, realised, …)
stochastic-rs-quant            pricing, calibration, vol surface, risk
stochastic-rs-ai               neural surrogates (feature-gated)
stochastic-rs-viz              Plotly grid plotter
stochastic-rs-py               PyO3 cdylib (210 entries)

Where to start

• New to the library? Quickstart — install, simulate a path, price an option, and read a Hurst estimate, end-to-end.
• Looking for a specific process / pricer? Use the catalog index: Processes, Distributions, Quant.
• Using Python? Start with Python bindings overview.

Navigation

The left sidebar is grouped into:

1. Start here — getting-started + concepts (traits, prelude, feature flags). Read this section once, end-to-end, before you dive into the catalog.
2. Reference — one page per process / distribution / pricer / calibrator / estimator / copula. Page templates are documented in the docs-writing SKILL.
3. Bindings — Python parity table and interop notes.
4. Guides — long-form tutorials (Heston calibration, fBm Hurst estimation, vol-surface pipeline, …) and benchmark dashboard.
5. Project — migration, contributing, API reference.