Research-to-industry capabilities
Technology
Four connected research tracks organized around quantitative workflows, with explicit assumptions, diagnostics and maturity levels.
Option Surface & Implied Risk Analytics
This track turns option quotes into structured, diagnosable market information.
| Capability | Research basis | Intended output | Initial boundary |
|---|---|---|---|
| Joint American call/put surface smoothing | Joint arbitrage-free smoothing | Adjusted price/IV surface, constraint diagnostics, quote-deviation report | Start from validated non-dividend settings |
| Implied-volatility inversion | Delta-family and bound methods | IV, residual, bounds, failure status | Shared computational primitive |
| Risk-neutral density & distribution | Risk-neutral density representations and Fourier-cosine methods | Density, CDF, skew and maturity comparisons | European-price input; tail and positivity checks |
| Option-implied quantiles | Model-free option-implied quantiles | Quantile curve, identifiable probability range, sensitivity | Clearly labeled under the risk-neutral measure |
Derivative Pricing & Risk Engine
A modular pricing stack built from continuous-time Markov chain approximations, transforms, semi-analytical methods and simulation.
Initial model and contract scope
- One reproducible base model plus one stochastic-volatility / stochastic-local-volatility model.
- European benchmark plus one discretely monitored barrier contract as a validated path-dependent example.
- Price, selected Greeks, parameter scenarios, numerical settings and explicit error states.
- Benchmark each method on a common hardware and accuracy target before making performance claims.
Longer-term expansion
Asian, Bermudan/American, occupation-time, timer, VIX/variance products, rough volatility and selected multi-asset claims can be added only as the corresponding implementations are validated.
Equity–Volatility Joint Calibration
The distinctive research direction here is simultaneous information extraction from the equity option market and the volatility-derivative market.
A flagship research preview can compare:
- a calibration using equity options only;
- a joint SPX/VIX calibration;
- errors for both markets separately;
- constraint residuals, runtime and solver failure flags.
The first public version should use precomputed market snapshots until sparse-data behavior, non-standard maturities and initialization are robust enough for controlled programmatic access.
Insurance & Annuity Analytics
Research on variable annuities and actuarial risk naturally forms a separate solution area because the contracts, data and validation criteria differ from trading-desk workflows.
Representative capabilities include VIX-linked fee design, GLWB/LTC valuation, multi-asset annuity Greeks, surrender analysis and risk measures. Any public example should fix the contract specification, mortality/health assumptions, surrender behavior and economic scenario set.