Deep Calibration of Market Simulations using Neural Density Estimators and Embedding Networks

ArXiv ID: 2311.11913 “View on arXiv”

Authors: Unknown

Abstract

The ability to construct a realistic simulator of financial exchanges, including reproducing the dynamics of the limit order book, can give insight into many counterfactual scenarios, such as a flash crash, a margin call, or changes in macroeconomic outlook. In recent years, agent-based models have been developed that reproduce many features of an exchange, as summarised by a set of stylised facts and statistics. However, the ability to calibrate simulators to a specific period of trading remains an open challenge. In this work, we develop a novel approach to the calibration of market simulators by leveraging recent advances in deep learning, specifically using neural density estimators and embedding networks. We demonstrate that our approach is able to correctly identify high probability parameter sets, both when applied to synthetic and historical data, and without reliance on manually selected or weighted ensembles of stylised facts.

Keywords: Agent-based simulation, Limit order book, Calibration, Neural density estimators, Deep learning, Equities (Microstructure)

Complexity vs Empirical Score

• Math Complexity: 7.5/10
• Empirical Rigor: 6.0/10
• Quadrant: Holy Grail
• Why: The paper involves advanced mathematical concepts like Bayesian inference, neural density estimators, and simulation-based inference with complex LaTeX notation, indicating high mathematical density. It demonstrates calibration on both synthetic and historical data with specific models (ZI and Chiarella), showing strong implementation and data requirements but stops short of full backtesting with trading strategies.

  flowchart TD
    A["Research Goal<br>Calibrate market simulators without<br>manual stylised fact weighting"] --> B["Data & Inputs<br>Simulated & Historical<br>Limit Order Book Data"]
    B --> C["Core Methodology<br>Deep Calibration with Neural Density<br>Estimators & Embedding Networks"]
    C --> D["Computational Process<br>Learn density of parameters<br>conditioned on observed data"]
    D --> E["Key Findings<br>Accurate identification of<br>high probability parameter sets<br>Validated on synthetic & historical data"]