Use Cases

Every quoted price is derived from a fitted volatility surface — if the surface has artifacts or arbitrage violations, the quotes are wrong, and better-informed counterparties will pick them off. Noise in fitted surfaces is particularly dangerous: it is indistinguishable from real signal, so the market maker cannot tell whether a fill reflects genuine flow or a fitting error. Systematic bias causes directional inventory accumulation. Getting the surface right is not alpha for market makers; it is table-stakes infrastructure that must work correctly at all times, across all market conditions.

Vola's Fitter produces surfaces under no-arbitrage constraints in real time. The C* curves are designed to avoid butterfly arbitrage (unlike SVI, which can produce negative probability densities), and calendar constraints are enforced across the full surface during fitting, not checked after the fact.

The Fitter's output error bars are particularly valuable here: they quantify uncertainty in each fitted vol, derived from input bid-ask spreads. For market makers, they serve as a natural "minimum edge" — if the market price is within the error band, there is no statistical confidence that edge exists. When data quality degrades (stale quotes, exchange glitches), error bars widen automatically, a graduated defense more sophisticated than hard kill switches. See fitted surface examples.

Relative value, skew trading, vol risk premium, event-driven: all of these strategies require a well-fitted surface as the reference against which mispricings are measured. Fitting artifacts generate spurious signals; traders cannot distinguish genuine edge from fitting noise.

Vola's Fitter and C* curves produce that reference surface, fitting all observed market shapes including W-shaped smiles around binary events that standard parameterizations cannot represent. The Pricer computes Greeks analytically from the parametric surface, including smart delta and gamma that account for how vol moves when spot moves.

For event-driven strategies around earnings, elections, and referenda, the Event Modeling and Event Var Fitter modules separate the clean (non-event) surface from discrete event components, isolating the risk being traded. The PnL Explanation module attributes P&L to spot, vol, time, and residual, showing whether returns come from the factors you intended to trade or from accidental exposure.

Options portfolios spanning hundreds of underliers need consistent Greeks — delta, gamma, vega, theta — for risk reporting, compliance, and hedge execution. Theta alone, expressed in dollars per day and aggregable across underliers and expiries, is often the single most important risk measure for any desk that runs a book overnight. When different underliers are fitted with different methods or tolerances, the result is phantom risk that no amount of downstream smoothing can fix.

Vola fits every underlier through the same methodology with no-arbitrage constraints, scaling to the entire US equity options universe. Vola's Pricer supports scenario analysis, shifting spot, vol level, skew, curvature, and other surface parameters to produce realistic stress tests. The PnL Explanation module attributes returns to spot, vol, time, and residual, showing whether hedges are performing as intended and surfacing uncaptured risks.

Structured products like autocallables, barrier notes, and cliquets are priced through local vol or stochastic local vol models that are calibrated from the vanilla vol surface. Butterfly arbitrage (negative implied density) or calendar arbitrage (negative forward variance) in the input surface propagates directly into downstream model calibration and pricing. Banks function as risk recyclers: packaging risk into structured products for institutional clients, then hedging it. The entire process depends on the quality of the vol surface at every stage.

Vola's Fitter enforces no-arbitrage constraints during fitting and extrapolates smoothly into the wings where listed options are illiquid or absent, producing the clean surfaces these models require. The Div Fitter handles discrete cash dividends correctly, essential for multi-year products on dividend-paying equities where continuous-yield approximations accumulate errors over the product lifetime. The Vol Derivatives module provides variance and volatility swap analytics derived from the fitted surface, ensuring consistency across vanilla and vol-derivative valuations.

Settlement prices for illiquid option series must come from a fitted surface, not observed trades. Margin models depend on Greeks from clean surfaces — noisy inputs mean either excess margin (capital inefficiency) or insufficient margin (counterparty risk).

Vola's C-curve families produce deterministic, reproducible fits with explicit quality metrics (weighted residuals, arbitrage margin diagnostics). The same library can be run independently by the exchange and its members, enabling transparent mark-to-model processes. The VIX Pricer computes VIX values directly from the fitted SPX surface, ensuring consistency between the underlier surface and derived index values.