Implied Volatility Surface & SVI Lab

Model institutional 3D volatility surfaces using Gatheral's SVI parameterization. Analyze cross-sectional smile skews, term structure contango/backwardation, and static calendar/butterfly no-arbitrage bounds.

3D Surface & SVI Geometry
Surface Presets:

SVI & Market Parameters

Slow (0.2) 1.2 Fast (3.0)
Gatheral SVI Skew Parameters
1M ATM Volatility

19.2%

30-Day Benchmark
1Y ATM Volatility

23.8%

365-Day Long-Term
Term Structure Regime

Contango

Slope: +4.6% vol
Arbitrage Status

Arbitrage-Free

Calendar & Butterfly OK
At-The-Money implied volatility progression across maturities from 7 days to 2 years, demonstrating mean reversion dynamics.
Cross-sectional volatility smiles across 1M, 3M, 6M, and 1Y maturities displaying skew flattening over longer horizons.
Total implied variance $w(k) = \sigma^2 T$ in log-moneyness coordinates $k = \ln(K/F)$, illustrating SVI asymptotic hyperbolic wings.

Implied Volatility Surface Matrix ($\sigma(K, T)$ %)

Sensitivity: Skew ($\rho$) vs. Maturity on 25-Delta Put Volatility (%)

Sensitivity: Wing Width ($b$) vs. Short-Term Vol on 90D OTM Call Vol (%)

Quantitative Foundations: The Volatility Surface & SVI Model

1. Anatomy of the Volatility Surface

In real financial markets, Black-Scholes implied volatilities vary across both strike price $K$ and expiration $T$. The volatility surface $\sigma(K, T)$ integrates:

  • The Smile / Skew (Strike Dimension): For equities, OTM puts trade at elevated implied volatilities due to market crash fears (leverage effect), producing a downward-sloping skew. For FX and crypto, both wings trade at a premium, producing a smile.
  • The Term Structure (Time Dimension): Near-term versus long-term volatility expectations driven by scheduled macroeconomic events, monetary policy cycles, and mean reversion.

2. Gatheral's Raw SVI Parameterization

Jim Gatheral (2004) proposed the Stochastic Volatility Inspired (SVI) formula for total implied variance $w(k) = \sigma^2(k) T$ as a function of log-moneyness $k = \ln(K/F)$:

$$w(k) = a + b \left[ \rho (k - m) + \sqrt{(k - m)^2 + \sigma^2} \right]$$

Where $a \in \mathbb{R}$ controls vertical level, $b \ge 0$ controls the angle of the hyperbolic asymptotes, $\rho \in (-1, 1)$ governs orientation and skewness, $m \in \mathbb{R}$ shifts the vertex horizontally, and $\sigma > 0$ controls vertex curvature.

3. Static No-Arbitrage Constraints

To be valid for risk-neutral pricing and exotic derivative simulation, the surface must satisfy two critical static conditions:

1. Calendar Spread Arbitrage-Free:
$$\frac{\partial w(k, T)}{\partial T} \ge 0 \quad \forall k \in \mathbb{R}$$ Total implied variance cannot decrease with maturity for identical log-moneyness strikes.
2. Butterfly Spread Arbitrage-Free (Convexity):
$$g(k) = \left(1 - \frac{k w'}{2 w}\right)^2 - \frac{w'^2}{4}\left(\frac{1}{w} + \frac{1}{4}\right) + \frac{w''}{2} \ge 0$$ Guarantees that the risk-neutral probability density function $q(K) \ge 0$ everywhere, preventing negative butterfly arbitrage payoffs.

Volatility Surface & SVI Mastery Quiz

1. In equity index markets, what sign does the SVI parameter $\rho$ typically take?
2. What constitutes a calendar spread arbitrage on a volatility surface?
3. What is total implied variance $w(k)$ in terms of annualized implied volatility $\sigma$?
4. How does volatility skew typically behave as time to expiration $T$ lengthens?

Related Volatility Modeling & Derivatives Simulators

Volatility Cone Lab Vanna-Volga Lab Breeden-Litzenberger Lab SABR Volatility Smile Lab Heston Stochastic Volatility Lab Variance Swap Lab Dispersion Trading Lab Risk Reversal Lab Synthetic Futures Lab Black-Scholes & Greeks Lab

Frequently Asked Questions

An Implied Volatility Surface is a three-dimensional geometric representation that maps an option's implied volatility as a function of two variables: strike price (or log-moneyness) and time to expiration (maturity). It synthesizes both the volatility smile/skew across strikes and the volatility term structure across time.

Introduced by Jim Gatheral in 2004, the SVI model parameterizes total implied variance $w(k) = \sigma^2 \cdot T$ as a function of log-moneyness $k$ using five intuitive parameters ($a, b, \rho, m, \sigma$). SVI guarantees consistent asymptotic behavior in the wings and facilitates quasi-explicit calibration free of static arbitrage.

An arbitrage-free surface must satisfy two primary static constraints: (1) Calendar spread arbitrage-free: total implied variance must be non-decreasing with expiration ($\partial w / \partial T \ge 0$); (2) Butterfly spread arbitrage-free: option prices must be convex in strike ($\partial^2 C / \partial K^2 \ge 0$), ensuring the extracted risk-neutral probability density is non-negative everywhere.

Contango occurs when near-term implied volatility is lower than long-term volatility, typical during calm bull markets. Backwardation occurs when short-term volatility spikes above long-term volatility, typical during market crashes, systemic liquidity crunches, or immediate pre-earnings uncertainty.

Log-moneyness $k$ is standardly defined relative to the forward price: $k = \ln(K / F) = \ln(K / S) - (r - q)T$. When $k = 0$, the strike equals the forward price (At-The-Money Forward). Negative $k$ values correspond to OTM puts, while positive $k$ values correspond to OTM calls.