SVI, SSVI, and SABR calibration for product systems.

Surface models matter most when their outputs can move into pricing, Greeks, scenarios, and spread workflows without translation.

Why this needs one shared risk path

!

Calibration routines are often isolated notebooks that never become product-grade infrastructure.

!

Model choice affects diagnostics, extrapolation behavior, and downstream risk.

!

Surface readouts need to be reproducible and inspectable.

How FerroRisk handles it

01

Select the model family

SVI, SSVI, and SABR are named explicitly as surface workflows.

02

Calibrate with quality gates

Fit diagnostics help distinguish usable surfaces from surfaces that should be rejected or reviewed.

03

Expose readouts

Smile panels and surface outputs can be consumed by applications, chain analytics, and scenario systems.

What makes this FerroRisk-shaped

Docs include dedicated surface calibration and smile panel use cases.

The engine is pure Rust and designed for production use.

Surface outputs sit beside pricing and risk primitives.

Common concerns

Why separate SVI, SSVI, and SABR?

Different products and markets need different surface model behavior. FerroRisk keeps model choice explicit.

Are diagnostics exposed?

Yes. Fit-quality diagnostics are part of the public FerroRisk positioning.