Explicit Black-Scholes Implied Volatility

• The paper introduces an explicit solution to compute Black-Scholes implied volatility via the inverse Gaussian quantile, eliminating iterative methods.
• It details alternative approaches including asymptotic expansions, Chebyshev interpolation, and Taylor series inversion to approximate implied volatility.
• The paper highlights computational advantages, demonstrating machine-precision accuracy and significant speedups compared to traditional Newton-Raphson solvers.

The Black-Scholes implied volatility is defined as the volatility parameter that, when input into the Black-Scholes formula, reproduces the observed market price of an option. For over fifty years, inversion of the Black-Scholes formula for implied volatility was regarded as a fundamentally transcendental problem: no explicit “closed-form” expression, in terms of elementary or classical special functions, was believed to exist. Recent advances have resolved key aspects of this problem by establishing both precise mathematical impossibility statements for traditional explicit forms and, more recently, by providing new explicit expressions involving special, non-classical functions. This article reviews the explicit solution to Black-Scholes implied volatility, its mathematical underpinnings, effective approximations, asymptotic expansions, and implications for computational practice.

1. Mathematical Formulation and Inversion Problem

Given the Black-Scholes formula for the value $C$ of a European call option: $C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$ where $S$ is spot, $K$ strike, $r$ risk-free rate, $T$ time to maturity, $\Phi$ the standard normal CDF, and

$$d_{1,2} = \frac{\ln(S/K) + (r \pm \tfrac{1}{2} \sigma^2)T}{\sigma\sqrt{T}}$$

implied volatility, $\sigma_{\mathrm{imp}}$, is defined as the unique positive root of

$$C_{\mathrm{BS}}(S, K, r, T, \sigma_{\mathrm{imp}}) = C^{\mathrm{market}}$$

Given $C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$0, all quantities are explicit except for $C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$1; the inversion is transcendental due to the nonlinear structure of the cumulative normal in $C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$2.

2. Impossibility of Traditional Closed-Form Expressions

Gerhold (2012) rigorously proved that the implied volatility map does not belong to the class of D-finite (holonomic) functions—those satisfying linear differential equations with polynomial coefficients—a broad class encompassing all algebraic, elementary, and classical special functions, including hypergeometric, Airy, and Bessel functions. The proof constructs a univariate specialization whose functional inverse exhibits singularity structure (fractional powers of logarithms) forbidden to D-finite functions, thereby excluding the existence of a closed form composed of such functions. No finite composition of hypergeometric or related expressions can yield the exact implied volatility as a function of option price, strike and spot (Gerhold, 2012).

A plausible implication is that all root-finding and iterative approximation routines, as well as rational or series approximants, are necessitated by the essential analytic complexity of the inversion unless non-classical function classes are permitted in the representation.

3. Explicit Solution via Inverse Gaussian Quantile

Schadner (2026) demonstrated that, by recasting the normalized Black-Scholes call price as a survival probability of the inverse Gaussian (IG) distribution, it is possible to express the implied volatility directly in terms of the inverse Gaussian quantile function.

Let

$C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$3

Define the normalized price

$C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$4

Then,

$C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$5

where $C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$6 denotes the IG survival function.

Inverting gives

$C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$7

where $C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$8 is the IG quantile. Recovering $C = S\,\Phi(d_1) - K\,e^{-rT}\,\Phi(d_2)$9: $S$0 with $S$1, $S$2 for $S$3, $S$4 for $S$5, and $S$6.

This formula is fully explicit: all terms depend only on observables and a single call to $S$7. No iterative procedure, no expansion, and no numerical integration are required. Modern libraries implement $S$8 efficiently and to machine precision (Schadner, 27 Apr 2026).

4. Asymptotic Expansions and Power Series Inversions

Prior to the explicit IG-quantile solution, several approaches constructed accurate closed-form approximations valid in specified regimes. Specifically, Grunspan (2011) derived:

• Asymptotic expansions for short-maturity ($S$9), large-maturity ($K$0), and extreme strikes ($K$1, $K$2), all based on transseries inversion of the Black-Scholes mapping; the first five terms of each expansion are provided explicitly with an inductive scheme for higher orders.
• An at-the-money power series in the call price, with coefficients defined recursively, convergent for $K$3.

These expansions afford explicit expressions in their domains and are valuable when extreme accuracy or performance is sought near the boundaries of practical option parameter regimes (Grunspan, 2011).

5. Explicit Approximation Approaches: Chebyshev Interpolation and Taylor Expansions

Where speed and accuracy are both required at scale, bivariate polynomial interpolation via Chebyshev expansions is effective. Glau, Grbac, Lučic, and Schwendner (2017) construct a tensor-product Chebyshev interpolant for the normalized inverse problem: $K$4 where $K$5, over the square $K$6 after suitable nonlinear transformations in $K$7 and $K$8 to ensure uniform accuracy. Given a prescribed tolerance, low-degree polynomials (built offline) yield near machine-precision results at evaluation speed up to 2.5x faster than optimized Newton-Raphson methods. The method is globally explicit, requiring only elementary arithmetic and evaluation of Chebyshev polynomials, and error decays sub-exponentially with degree (Glau et al., 2017).

In the context of local-stochastic volatility models, Lorig, Pagliarani, and Pascucci (2013) described explicit implied volatility expansions by Taylor expansion of the price map and term-by-term analytic inversion. Each coefficient in the expansion is explicitly expressed in terms of model coefficients, derivatives of the Black-Scholes formula, and Hermite polynomials. The resulting approximations rapidly converge for short maturities and are computationally inexpensive (Lorig et al., 2013).

6. Computational Properties and Performance

The explicit IG-quantile formula provides a single-step computation of implied volatility that attains double-precision accuracy across the full practical range of parameters. In benchmarking, the mean absolute error is on the order of $K$9 (machine precision) with a per-evaluation speedup of 3.4x over the “Let’s Be Rational” Jäckel benchmark (Schadner, 27 Apr 2026). Chebyshev polynomial interpolants likewise achieve machine-precision across broad domains with 0.4–0.7x the time of Newton-based solvers for batch evaluation (Glau et al., 2017).

| Method | Max |Δσ| | Mean |Δσ| | Relative Runtime | |------------------------------------------|----------|------------|-------------------| | Explicit IG-Quantile (Schadner, 27 Apr 2026) | 2.2e-16 | 2.2e-16 | 0.3 | | Chebyshev (high-acc) (Glau et al., 2017) | 5e-11 | 5e-12 | 0.67 | | Newton–Raphson (reference) | 2e-10 | 6e-14 | 1 |

Runtime is normalized to Newton–Raphson as 1.0; Max and Mean |Δσ| are maximum and mean absolute errors across a large grid.

7. Implications and Open Questions

The analytic structure of Black-Scholes implied volatility precludes closed-form expressions in terms of elementary and most classical special functions, as established by Gerhold’s non-holonomicity result (Gerhold, 2012). The new explicit formula built on the inverse Gaussian quantile circumvents this constraint by appealing to non-D-finite functions outside the classical hierarchy.

In practice, both the IG-quantile solution and Chebyshev polynomial approximations render iterative root-finding obsolete, reducing both latency and error in large-scale pricing engines. Open research questions remain regarding further characterizing the analytic class of implied volatility, extension to other models (e.g., stochastic volatility, jump-diffusion), and the discovery of alternative explicit representations potentially involving other nonlinear special functions. The hierarchy of asymptotic expansions and polynomial approximants provides robust practical solutions for regimes where the quantile representation is less convenient or where parametric flexibility for accuracy–speed tradeoff is desired.