Towards Calibrating Financial Market Simulators with High-frequency Data
Peng Yang, Junji Ren, Feng Wang, Ke Tang
TL;DR
Calibrating financial market simulators at high frequency is hampered by non-identifiability and information loss when using traditional discrepancy measures. The authors advocate a KS-based objective to preserve high-frequency detail and analyze its identifiability, finding improved parameter identifiability at high frequencies but increased multi-modality. To navigate this landscape, they develop an Improved Negatively Correlated Search (NCS) with Adaptive Stochastic Ranking that balances exploration and exploitation across multiple Gaussian search processes. Empirical results on a PGPS-based market simulator show KS calibration yields higher fidelity than MSM, with synthetic data improvements up to about 36% and real data gains up to roughly 16%, albeit with substantial computational costs. The work points to promising directions for efficient, high-fidelity, high-frequency calibration and highlights the need for scalable methods across broader data regimes.
Abstract
The fidelity of financial market simulation is restricted by the so-called "non-identifiability" difficulty when calibrating high-frequency data. This paper first analyzes the inherent loss of data information in this difficulty, and proposes to use the Kolmogorov-Smirnov test (K-S) as the objective function for high-frequency calibration. Empirical studies verify that K-S has better identifiability of calibrating high-frequency data, while also leads to a much harder multi-modal landscape in the calibration space. To this end, we propose the adaptive stochastic ranking based negatively correlated search algorithm for improving the balance between exploration and exploitation. Experimental results on both simulated data and real market data demonstrate that the proposed method can obtain up to 36.0% improvement in high-frequency data calibration problems over the compared methods.