Skip to content
VolSurface Desk
Directory

Svi Sabr Smile Arbitrage Free

Natural SVI and SVI-Jump-Wings Parameter Mapping for Equity Options Desks

SVI-Jump-Wings (SVI-JW) bijectively maps abstract hyperbolic coordinates into five observable trader metrics—ATM variance, ATM skew, put/call wing slopes, and minimum variance.

By Dr. Alistair Finch, CQF, FRMReview not recorded
Natural SVI and SVI-Jump-Wings Parameter Mapping for Equity Options Desks — VolSurface Desk 2026 quantitative & supervisory data card
VolSurface Desk (SVI & SABR Volatility Surface Calibration & No-Arbitrage Conditions): Trader-native ATM variance, skew, and jump-wing slope coordinates mapped bijectively to raw SVI.Source: VolSurface Desk Institutional Research DeskLicense: CC-BY-4.0

Market makers quoting single-stock and index equity options never adjust raw hyperbolic shift a or apex smoothing σ manually during live trading; instead, institutional Volatility Marking Terminals expose Gatheral's SVI-Jump-Wings (SVI-JW) coordinates (vt, ψt, pt, ct, ṽ_t), which remain time-scale invariant across expirations from 7 days to 2 years. Built as an exact mathematical bijection over Gatheral raw SVI five-parameter total variance fitting and unified within our SVI & SABR Volatility Surface Calibration & No-Arbitrage Conditions guide, Natural SVI and SVI-JW allow desk traders to bump ATM skew by +0.50 vol points per 10% moneyness without distorting ATM implied volatility.

Three Equivalent Representations of the Same Smile#

Gatheral and Jacquier (2014) established exact algebraic equivalence across three SVI coordinate systems:

  1. Raw SVI (a, b, ρ, m, σ): Optimal for fast Zeliade quadratic programming and analytical differentiation.
  2. Natural SVI (Δ, μ, ρ, ω, ζ): Centers the parameterization around ATM variance scale ω ≥ 0 and curvature ζ > 0 via w(k) = Δ + (ω / 2) × [1 + ζρ(k − μ) + √((ζ(k − μ) + ρ)2 + (1 − ρ2))], cleanly extending into Surface SVI power-law kernels when Δ = 0 and μ = 0.
  3. SVI-Jump-Wings (vt, ψt, pt, ct, ṽ_t): Scales every coordinate by √w(0) or 1 / T so values match annualized variance and normalized Black-Scholes volatility slopes directly observable on trading screens alongside our Implied Volatility Calculator.

Bijective Coordinate Mapping: Raw SVI <-> Natural SVI <-> SVI-JW#

Bijective transformation diagram between Raw SVI Natural SVI and SVI-Jump-Wings trader coordinates
Bijective transformation diagram between Raw SVI Natural SVI and SVI-Jump-Wings trader coordinatesVerified Exhibit

Exact Bijection Equations and Inverse Recovery#

Given calibrated raw SVI parameters (a, b, ρ, m, σ) at time-to-expiry T, let w_t = w(0) denote total ATM variance. The five SVI-JW coordinates are defined analytically—and invert uniquely back to raw SVI whenever m ≠ 0:

Formula or model from the articleNot independently verified
v_t = w_t / T, ψ_t = (b / 2√w_t)·(−m / √(m2 + σ2) + ρ), p_t = b(1 − ρ)/√w_t, c_t = b(1 + ρ)/√w_t, ṽ_t = [a + bσ√(1 − ρ2)] / T
  • Inverse Map (SVI-JW -> Raw SVI): b = (√wt / 2)·(ct + pt), ρ = 1 − (pt·√wt)/b, β = ρ − (2·ψt·√w_t)/b (requires |β| ≤ 1).
  • Worked NVDA 30-DTE Equity Option Slice (T = 0.0822 yr): trader marks √vt = 44.0% (wt = 0.01591), ψt = −0.38, pt = 0.92, ct = 0.48, √ṽt = 41.8%.
  • Step-by-Step Recovery: b = (0.12614 / 2) × 1.40 = 0.08830; ρ = 1 − (0.92 × 0.12614)/0.08830 = −0.3143; β = −0.3143 + 1.0857 = +0.7714.
  • Vertex & Shift Output: α = sign(β)·√(1/β2 − 1) = +0.8248 => m = +0.0412, σ = 0.03398, a = 0.01153 (0.04 ms exact inversion).

Notice the geometric compatibility test |β| ≤ 1: the inverse exists smoothly if and only if ATM skew ψt lies strictly between the left put wing slope −pt / 2 and right call wing slope +c_t / 2.

Cross-Sectional Stability Across Equity vs Index Names#

Why do institutional desks store historical end-of-day smile marks in SVI-JW rather than raw SVI? Across expirations, raw parameter b grows roughly as √T, making cross-tenor comparison awkward. Dividing by √wt in SVI-JW stabilizes (pt, ct, ψt) across 1-month, 3-month, and 6-month expiries so traders immediately spot term-structure kinks or Vomma wing dislocations tracked in Third-Order Greeks: Vomma, Zomma & Speed and our Option Greeks & GEX Calculator.

SVI-Jump-Wings parameter comparison across SPX Index Nasdaq-100 Mega-Cap Tech and Biotech Equity Options
SVI-Jump-Wings parameter comparison across SPX Index Nasdaq-100 Mega-Cap Tech and Biotech Equity OptionsVerified Exhibit
Quantitative Distribution · SPX 30D Benchmark
SPX 30D BenchmarkSingle-Stock Tech 30D
01√v_t (ATM IV)17.80%02p_t (Put Wing)1.4803c_t (Call Wing)0.2204√ṽ_t (Min IV)15.95%
CoordinateFormula / ScaleSPX 30D BenchmarkSingle-Stock Tech 30DDesk Hedging / Risk Link
√v_t (ATM IV)√(w(0) / T)17.80%38.50%Black-Scholes Calculator
ψ_t (ATM Skew)w'(0) / (2√w_t)−0.64−0.26SPX 25-Delta Skew
p_t (Put Wing)b(1 − ρ) / √w_t1.480.88SABR-SVI Surface Standard
c_t (Call Wing)b(1 + ρ) / √w_t0.220.54Cboe Listed Venue
√ṽ_t (Min IV)√(w_min / T)15.95%36.90%Vomma & Zomma Mechanics

Butterfly Safety When Manually Bumping SVI-JW Wings#

When a trader manually raises pt or ct during a market selloff on an NFA-supervised derivatives desk, raising wing slopes while holding vt − ṽt too tight can violate Durrleman butterfly positivity. Applying Gatheral-Jacquier's sufficient condition—setting vt and ψt from market ATM quotes while floor-clamping minimum variance via ṽt ≥ vt × 4·pt·ct / (pt + ct)2—provides an instant automated guardrail inside SVI & SABR Volatility Surface Calibration & No-Arbitrage Conditions before quotes hit the wire.

Frequently asked questions

What happens in the SVI-JW inverse transformation when m equals exactly zero?

When m = 0, the hyperbola minimum sits exactly at the forward (k = 0), so v_tilde_t equals v_t and ATM skew psi_t equals b*rho/(2*sqrt(w_t)); the limiting algebraic branch sets m = 0 directly and solves sigma = sqrt(w_t) / b without dividing by beta.

Why is SVI-JW preferred over 25-delta risk reversals and butterflies for equity options?

Black-Scholes delta depends on implied volatility itself, requiring iterative root-finding to map 25-delta strikes back to log-moneyness, whereas SVI-JW is defined directly at k = 0 and asymptotic |k| -> infinity with a closed-form bijection.

How does Natural SVI differ from SVI-Jump-Wings?

Natural SVI (Delta, mu, rho, omega, zeta) parameterizes total variance around the smile minimum location mu and scale omega, serving as the theoretical bridge to global Surface SVI (SSVI), whereas SVI-JW anchors directly at ATM forward k = 0 for live desk marking.

Can a desk interpolate SVI-JW coordinates linearly across adjacent option expirations?

Linearly interpolating total ATM variance w_t = v_t * T alongside bounded SVI-JW weights between adjacent slices preserves butterfly positivity much more reliably than interpolating raw SVI parameters (a, b, rho, m, sigma) directly.

Publication record

Published
November 14, 2026
Updated
November 14, 2026
Author
Dr. Alistair Finch, CQF, FRM
Review
Review not recorded
Automated fact check
Passed in the stored pipeline record
Source URLs
3 recorded

AI assistance recorded: anthropic/claude-opus-4.6. This field does not establish human review or source verification.

Academic & Institutional Attribution

Cite This Research & Embed Live Data Card

Open Attribution License (CC BY 4.0)

Standard academic & institutional research paper citation format.

Dr. Alistair Finch, CQF, FRM. (2026, November 14). Natural SVI and SVI-Jump-Wings Parameter Mapping for Equity Options Desks. VolSurface Desk. https://volsurfacedesk.com/desk-notes/natural-svi-jump-wings-jw-trader-parameter-mapping
Explore Svi Sabr Smile Arbitrage Free Hub (20) →

VolSurface Desk Data, Benchmarks & Contributor Records

Natural SVI & Jump-Wings (SVI-JW) Parameter Mapping | VolSurface Desk