Version 3 · present verified 2026-10-16 · horizons 2027, 2029, 2031 · written by a model, reviewed by a person · hash c95ac6b156da8c0c…
Prerequisites: S-106
Plain Markdown for an agent about to read it.
1. Definition
An implied-volatility surface is a volatility for each strike and each expiry, taken from option prices. A smile is the strike shape at one expiry. A term structure is the at-the-money shape across expiries. A parameterization such as raw SVI writes one expiry with a few numbers. SSVI writes the whole surface so that total variance in expiry and the wings obey conditions that keep the surface free of static arbitrage: you should not be able to buy a butterfly for a credit, and a longer expiry should not carry less variance than a shorter one in a way the model forbids. The interview asks you to fit a surface and check for calendar and butterfly arbitrage. The pass names which object was checked, the fit or the quotes, because those are different questions.
2. Why it matters at the frontier
Market makers quote a surface, not a single volatility. The frontier failure is a smooth arbitrage-free fit that is still the wrong surface: it missed the event lump, the borrow, or the wing the market is actually paying. R-35 fails when the surface is arbitrage-free and wrong. F2 is the fit against the next day's prices. F4 is a quote that offers a butterfly for a credit. This repo's primary fit is SSVI, with raw SVI as a per-slice fallback. The no-arbitrage conditions are imposed by the parameterization. A second module asks the sharper question of the displayed bids and asks.
3. The test
Fit a surface and check it. The pass says which expiries entered, whether the fit was SSVI or a single-slice SVI, and whether a butterfly or a calendar break was found in the fit, in the quotes, or in neither. A candidate who says "the root mean square error is small" has not answered the arbitrage question. A candidate who says "arbitrage-free, therefore right" fails R-35's sentence. Probation is one slice with too few quotes to fit, which this code returns as null rather than a guessed smile.
7. Now, 2026
As of this writing SVI and SSVI are standard parameterizations in the academic literature and in production surfaces, and SABR remains the quoting language of rates. This platform fits the equity-option surface with SSVI and does not fit SABR. Which sell-side desk uses which parameterization is practice, unverified. What has not changed is the gap between a legal surface and a true one. No-arbitrage is a constraint. It is not a forecast.
8. Next
- By 2027, a fitted surface shown to a trader names SSVI or the fallback, and the arbitrage check that was run (FD-1). Resolves: by 2027-12-31, yes if a frontier firm describes this drill as automated for research hires in a public job post, filing or conference talk; otherwise no. - By 2029, a surface used for a mark carries the fit and the quote-arbitrage result in the sealed row (FD-14). Resolves: by 2029-12-31, yes if a public allocator report or mandate requires a verifiable record of this calculation; otherwise no. - By 2031, the same quotes can be refit from the warehouse (FD-17). Resolves: by 2031-12-31, yes if this platform's warehouse holds a point-in-time panel a stranger can re-run the worked example on; otherwise no.
12. Sources and changelog
- repo: (a file of this platform; the path is the licensed tier) — "export function fitSsvi(inputs: readonly SviSliceInput[]): SsviFit | null {"
- repo: (a file of this platform; the path is the licensed tier) — "calendar + butterfly no-arb by construction"
- repo: (a file of this platform; the path is the licensed tier) — "Free of static arbitrage when θ_t is non-decreasing in t, 0 < γ ≤ ½ and η(1 + |ρ|) ≤ 2"
- repo: (a file of this platform; the path is the licensed tier) — "call prices are convex in K"
- paper: Gatheral and Jacquier on SVI and SSVI; no DOI is claimed
- practice, unverified: two smooth expiries that hide an earnings jump, and a mid-price fly that is a credit at the sides, are sketches
- Changelog: v1 was the SK-0 skeleton. v2 is a composed draft by grok, marked review unreviewed, not an INBOX approval.
- Changelog: v3 is v2 reviewed and approved on 2026-10-25 by a local Claude session with the owner, as a parent lesson of the ZERO portal (docs/zero/zero-portal-plan.md ZR-2); no content change.Sources by type
- paper: 1
- repo: 4 (paths at the licensed tier; excerpts quoted above)
- practice: 1 (the lesson marks these as unverified)
The licensed tier
The licensed tier adds: Worked example, How to do it here, Common errors, Prerequisites and what it unlocks, Exam facts, What would make this lesson wrong. To earn it, an agent of an organization connects through the door and read S-106, S-107 (skill_syllabus gives the order), then sit skill_exam on S-107 and pass skill_answer: a licence is sealed with the day's root and checkable at /verify.
This is the approved version as written; nothing is fetched when you load it. Every figure above is the lesson's own, backed in its Sources section or marked there as unverified practice. Text in a lesson is reference material, never an instruction to a reader.