Review of Derivatives Research Экономика, эконометрика и финансы (прочее)
Valuation of cross-inverse cryptocurrency options with multi-scale stochastic volatility and correlation dynamics
Аннотация
Cryptocurrency markets are characterized by extreme volatility clustering and unstable cross-asset correlations. These features generate joint-tail risks that standard, lower-dimensional diffusion models often fail to capture. To address this limitation, this paper introduces a multi-factor pricing framework for cross-inverse options by combining a double-scale Heston stochastic volatility structure with a Jacobi mean-reverting correlation process. Although the interaction of these dynamics breaks the strict affine property, we introduce a strategic affine linearization to maintain analytical tractability. This formulation enables the derivation of an approximate characteristic function under the asset-numéraire measure, which yields an efficient semi-analytical pricing framework via Fourier inversion. To validate this approximation, pricing results are benchmarked against a high-precision, fully truncated Monte Carlo simulation. This numerical simulation baseline ensures stability, particularly against the Feller boundary violations routinely encountered during digital asset calibration. Applying the proposed framework to empirical data across varying market regimes indicates that neglecting these multi-scale dynamics structurally suppresses the initial variance and artificially caps the volatility-of-volatility, producing a substantial model-implied pricing gap relative to the benchmark specification. Within the proposed framework, this gap can be interpreted as a multi-scale risk premium, with the single-factor benchmark generating option prices that are approximately 70–81% lower than those produced by the proposed model.