CaseMark Analyzing Exotic Options
When To Use
- Pricing or valuing barrier options (knock-in, knock-out, single/double barrier)
- Analyzing Asian (average-price or average-strike) options for hedging or structured product design
- Evaluating lookback options (floating or fixed strike) for optimal entry/exit analysis
- Pricing digital (binary/cash-or-nothing) options for event-driven payoffs
- Structuring exotic payoffs within structured notes, CLNs, or autocallable products
- Comparing exotic option economics against vanilla alternatives for a given hedge or exposure
Inputs To Gather
- Underlying specification: asset class, ticker/identifier, spot price, dividend/carry assumptions
- Option type and subtype: barrier (up-and-in, down-and-out, etc.), Asian (arithmetic vs. geometric, discrete vs. continuous averaging), lookback (floating vs. fixed strike), digital (cash-or-nothing vs. asset-or-nothing)
- Strike, barrier levels, and observation conventions: barrier monitoring frequency (continuous vs. discrete), averaging dates/windows, lookback observation period
- Expiry and exercise style: European, American, Bermudan; early exercise provisions if any
- Market data: volatility surface (skew/term structure), risk-free rate curve, correlation matrix (for multi-asset exotics), repo/borrow rates
- Model preferences: Black-Scholes analytic (where closed-form exists), Monte Carlo (path-dependent), PDE/finite-difference, tree-based methods
- Notional, currency, and settlement terms
- Purpose: trading desk pricing, risk management hedge analysis, structured product design, or model validation
Workflow
Classify the exotic structure
- Identify the payoff type: path-dependent (Asian, lookback, barrier) vs. event-triggered (digital)
- Map the exact payoff formula — express mathematically before coding or modeling
- Determine whether closed-form solutions exist (e.g., Merton for continuous barriers, Rubinstein for lookbacks) or if simulation is required
Select and configure the pricing model
- For continuous-barrier Europeans with lognormal dynamics: use analytic formulas (Reiner-Rubinstein for single barriers) [VERIFY applicability to discrete monitoring]
- For discrete barriers: apply Broadie-Glasserman-Kou continuity correction or use Monte Carlo with fine time-stepping
- For arithmetic Asian options: no exact closed-form — use Monte Carlo with variance reduction (control variate using geometric Asian) or moment-matching approximations (Turnbull-Wakeman, Levy)
- For lookback options: use analytic formulas for continuous observation (Goldman-Sosin-Gatto for floating strike) [VERIFY whether observation is truly continuous or discrete]
- For digitals: Black-Scholes closed-form with spread approximation for risk management; adjust for skew using call/put spread replication
- Specify simulation parameters if Monte Carlo: number of paths (minimum 100k for convergence), time steps (align to observation dates), random number generation (Sobol recommended for path-dependents), seed for reproducibility
Calibrate inputs and run sensitivities
- Fit the volatility surface — exotic prices are highly sensitive to skew and term structure; use market-implied vols at relevant strikes/tenors, not ATM flat vol
- For barrier options: compute vanna and volga adjustments if using a local/stochastic vol model [VERIFY model choice aligns with desk conventions]
- Run Greeks: delta, gamma, vega, theta; for barriers also compute barrier delta (sensitivity to barrier level) and gap risk near knock-out
- For Asians: assess the impact of fixing schedule changes and partial averaging (options mid-averaging period)
- Stress test: shift vol surface, spot, rates, and correlation; identify largest P&L drivers
Analyze path-dependency and risk characteristics
- For barriers: quantify pin risk near the barrier at expiry, assess knock-in vs. knock-out economics, evaluate rebate value
- For Asians: compare arithmetic vs. geometric pricing gap, assess vol reduction from averaging, evaluate Greeks evolution as fixings accumulate
- For lookbacks: analyze the premium over vanilla (lookback premium), assess whether cost is justified by the hedge benefit
- For digitals: quantify discontinuity risk at the strike, evaluate overhedge cost of call/put spread replication at various widths
Benchmark and validate
- Cross-check analytic prices against Monte Carlo (where both are available) — differences should be within 2-3 standard errors
- Compare against Bloomberg DLIB, FinCAD, or internal library pricing [VERIFY available systems]
- Validate Greeks numerically (bump-and-reprice) against analytic Greeks
- If structured product: decompose into component exotics and vanilla pieces, verify sum matches full-structure price
Document findings
- State the pricing model used, key assumptions (vol surface, dividend treatment, barrier monitoring convention), and limitations
- Present fair value, Greeks, and scenario analysis in tabular form
- Highlight model risk: where does the chosen model likely under/overstate value (e.g., local vol vs. stochastic vol for barriers)
- Flag any approximations and their estimated impact
Output
- Exotic option valuation summary: fair value (mid), bid/ask adjustment methodology, and confidence interval (for MC-based prices)
- Greeks table: delta, gamma, vega, theta, rho; plus exotic-specific sensitivities (barrier delta, fixing sensitivity for Asians)
- Scenario matrix: P&L under spot/vol/rate shifts and correlation stress
- Model comparison (if applicable): side-by-side pricing from analytic vs. MC vs. alternative models with variance explained
- Risk commentary: key risks (pin risk, gap risk, model risk, discrete monitoring bias), hedging recommendations, and replication cost estimates
- Payoff diagram: mathematical payoff specification and, where useful, illustrative payoff profiles at expiry
Quality Checks
- Payoff formula matches the term sheet or trade confirmation exactly — verify barrier direction, observation convention, averaging methodology, and settlement type
- Monte Carlo standard error is below 0.5% of option value; increase paths or apply variance reduction if not
- Greeks are internally consistent: delta integrates to option value across spot range, put-call parity holds for European-style exotics where applicable
- Barrier option price converges to vanilla price as barrier moves far from spot (knock-in) or to zero as barrier approaches spot (knock-out)
- Asian option price is bounded between the geometric average option price (lower) and the vanilla option price (upper) for calls
- Digital option price equals the negative of the derivative of the vanilla call price with respect to strike — verify numerically
- All jurisdiction-specific regulatory constraints on exotic sales or structured product classification are noted [VERIFY: MiFID II target market, Dodd-Frank SEF eligibility, ISDA documentation requirements]