Neural Network

Exact Terminal Condition Neural Network for American Option Pricing Based on the Black-Scholes-Merton Equations ArXiv ID: 2510.27132 “View on arXiv” Authors: Wenxuan Zhang, Yixiao Guo, Benzhuo Lu Abstract This paper proposes the Exact Terminal Condition Neural Network (ETCNN), a deep learning framework for accurately pricing American options by solving the Black-Scholes-Merton (BSM) equations. The ETCNN incorporates carefully designed functions that ensure the numerical solution not only exactly satisfies the terminal condition of the BSM equations but also matches the non-smooth and singular behavior of the option price near expiration. This method effectively addresses the challenges posed by the inequality constraints in the BSM equations and can be easily extended to high-dimensional scenarios. Additionally, input normalization is employed to maintain the homogeneity. Multiple experiments are conducted to demonstrate that the proposed method achieves high accuracy and exhibits robustness across various situations, outperforming both traditional numerical methods and other machine learning approaches. ...

End-to-End Large Portfolio Optimization for Variance Minimization with Neural Networks through Covariance Cleaning ArXiv ID: 2507.01918 “View on arXiv” Authors: Christian Bongiorno, Efstratios Manolakis, Rosario Nunzio Mantegna Abstract We develop a rotation-invariant neural network that provides the global minimum-variance portfolio by jointly learning how to lag-transform historical returns and how to regularise both the eigenvalues and the marginal volatilities of large equity covariance matrices. This explicit mathematical mapping offers clear interpretability of each module’s role, so the model cannot be regarded as a pure black-box. The architecture mirrors the analytical form of the global minimum-variance solution yet remains agnostic to dimension, so a single model can be calibrated on panels of a few hundred stocks and applied, without retraining, to one thousand US equities-a cross-sectional jump that demonstrates robust out-of-sample generalisation. The loss function is the future realized minimum portfolio variance and is optimized end-to-end on real daily returns. In out-of-sample tests from January 2000 to December 2024 the estimator delivers systematically lower realised volatility, smaller maximum drawdowns, and higher Sharpe ratios than the best analytical competitors, including state-of-the-art non-linear shrinkage. Furthermore, although the model is trained end-to-end to produce an unconstrained (long-short) minimum-variance portfolio, we show that its learned covariance representation can be used in general optimizers under long-only constraints with virtually no loss in its performance advantage over competing estimators. These gains persist when the strategy is executed under a highly realistic implementation framework that models market orders at the auctions, empirical slippage, exchange fees, and financing charges for leverage, and they remain stable during episodes of acute market stress. ...

Deep Learning Meets Queue-Reactive: A Framework for Realistic Limit Order Book Simulation ArXiv ID: 2501.08822 “View on arXiv” Authors: Unknown Abstract The Queue-Reactive model introduced by Huang et al. (2015) has become a standard tool for limit order book modeling, widely adopted by both researchers and practitioners for its simplicity and effectiveness. We present the Multidimensional Deep Queue-Reactive (MDQR) model, which extends this framework in three ways: it relaxes the assumption of queue independence, enriches the state space with market features, and models the distribution of order sizes. Through a neural network architecture, the model learns complex dependencies between different price levels and adapts to varying market conditions, while preserving the interpretable point-process foundation of the original framework. Using data from the Bund futures market, we show that MDQR captures key market properties including the square-root law of market impact, cross-queue correlations, and realistic order size patterns. The model demonstrates particular strength in reproducing both conditional and stationary distributions of order sizes, as well as various stylized facts of market microstructure. The model achieves this while maintaining the computational efficiency needed for practical applications such as strategy development through reinforcement learning or realistic backtesting. ...

Interpretable ML for High-Frequency Execution ArXiv ID: 2307.04863 “View on arXiv” Authors: Unknown Abstract Order placement tactics play a crucial role in high-frequency trading algorithms and their design is based on understanding the dynamics of the order book. Using high quality high-frequency data and a set of microstructural features, we exhibit strong state dependence properties of the fill probability function. We train a neural network to infer the fill probability function for a fixed horizon. Since we aim at providing a high-frequency execution framework, we use a simple architecture. A weighting method is applied to the loss function such that the model learns from censored data. By comparing numerical results obtained on both digital asset centralized exchanges (CEXs) and stock markets, we are able to analyze dissimilarities between feature importances of the fill probability of small tick crypto pairs and Euronext equities. The practical use of this model is illustrated with a fixed time horizon execution problem in which both the decision to post a limit order or to immediately execute and the optimal distance of placement are characterized. We discuss the importance of accurately estimating the clean-up cost that occurs in the case of a non-execution and we show it can be well approximated by a smooth function of market features. We finally assess the performance of our model with a backtesting approach that avoids the insertion of hypothetical orders and makes possible to test the order placement algorithm with orders that realistically impact the price formation process. ...