Towards Realistic and Interpretable Market Simulations: Factorizing Financial Power Law using Optimal Transport ArXiv ID: 2507.09863 “View on arXiv” Authors: Ryuji Hashimoto, Kiyoshi Izumi Abstract We investigate the mechanisms behind the power-law distribution of stock returns using artificial market simulations. While traditional financial theory assumes Gaussian price fluctuations, empirical studies consistently show that the tails of return distributions follow a power law. Previous research has proposed hypotheses for this phenomenon – some attributing it to investor behavior, others to institutional demand imbalances. However, these factors have rarely been modeled together to assess their individual and joint contributions. The complexity of real financial markets complicates the isolation of the contribution of a single component using existing data. To address this, we construct artificial markets and conduct controlled experiments using optimal transport (OT) as a quantitative similarity measure. Our proposed framework incrementally introduces behavioral components into the agent models, allowing us to compare each simulation output with empirical data via OT distances. The results highlight that informational effect of prices plays a dominant role in reproducing power-law behavior and that multiple components interact synergistically to amplify this effect. ...

Boltzmann Price: Toward Understanding the Fair Price in High-Frequency Markets ArXiv ID: 2507.09734 “View on arXiv” Authors: Przemysław Rola Abstract In this paper, we introduce a parametrized family of prices derived from the Maximum Entropy Principle. The price is obtained from the distribution that minimizes bias, given the bid and ask volume imbalance at the top of the order book. Under specific parameter choices, it closely approximates the mid-price or the weighted mid-price. Using probabilities of bid and ask states, we propose a model of price dynamics in which both drift and volatility are driven by volume imbalance. Compared to standard models like Bachelier or Geometric Brownian Motion with constant volatility, our model can generate higher kurtosis and heavy-tailed distributions. Additionally, the drift term naturally emerges as a consequence of the order book imbalance. We validate the model through simulation and demonstrate its fit to historical equity data. The model provides a theoretical framework, integrating price, volume imbalance, and spread. ...

Enhancing Trading Performance Through Sentiment Analysis with Large Language Models: Evidence from the S&P 500 ArXiv ID: 2507.09739 “View on arXiv” Authors: Haojie Liu, Zihan Lin, Randall R. Rojas Abstract This study integrates real-time sentiment analysis from financial news, GPT-2 and FinBERT, with technical indicators and time-series models like ARIMA and ETS to optimize S&P 500 trading strategies. By merging sentiment data with momentum and trend-based metrics, including a benchmark buy-and-hold and sentiment-based approach, is evaluated through assets values and returns. Results show that combining sentiment-driven insights with traditional models improves trading performance, offering a more dynamic approach to stock trading that adapts to market changes in volatile environments. ...

Mapping Crisis-Driven Market Dynamics: A Transfer Entropy and Kramers-Moyal Approach to Financial Networks ArXiv ID: 2507.09554 “View on arXiv” Authors: Pouriya Khalilian, Amirhossein N. Golestani, Mohammad Eslamifar, Mostafa T. Firouzjaee, Javad T. Firouzjaee Abstract Financial markets are dynamic, interconnected systems where local shocks can trigger widespread instability, challenging portfolio managers and policymakers. Traditional correlation analysis often miss the directionality and temporal dynamics of information flow. To address this, we present a unified framework integrating Transfer Entropy (TE) and the N-dimensional Kramers-Moyal (KM) expansion to map static and time-resolved coupling among four major indices: Nasdaq Composite (^IXIC), WTI crude oil (WTI), gold (GC=F), and the US Dollar Index (DX-Y.NYB). TE captures directional information flow. KM models non-linear stochastic dynamics, revealing interactions often overlooked by linear methods. Using daily data from August 11, 2014, to September 8, 2024, we compute returns, confirm non-stationary using a conduct sliding-window TE and KM analyses. We find that during the COVID-19 pandemic (March-June 2020) and the Russia-Ukraine crisis (Feb-Apr 2022), average TE increases by 35% and 28%, respectively, indicating heightened directional flow. Drift coefficients highlight gold-dollar interactions as a persistent safe-haven channel, while oil-equity linkages show regime shifts, weakening under stress and rebounding quickly. Our results expose the shortcomings of linear measures and underscore the value of combining information-theoretic and stochastic drift methods. This approach offers actionable insights for adaptive hedging and informs macro-prudential policy by revealing the evolving architecture of systemic risk. ...

MountainLion: A Multi-Modal LLM-Based Agent System for Interpretable and Adaptive Financial Trading ArXiv ID: 2507.20474 “View on arXiv” Authors: Siyi Wu, Junqiao Wang, Zhaoyang Guan, Leyi Zhao, Xinyuan Song, Xinyu Ying, Dexu Yu, Jinhao Wang, Hanlin Zhang, Michele Pak, Yangfan He, Yi Xin, Jianhui Wang, Tianyu Shi Abstract Cryptocurrency trading is a challenging task requiring the integration of heterogeneous data from multiple modalities. Traditional deep learning and reinforcement learning approaches typically demand large training datasets and encode diverse inputs into numerical representations, often at the cost of interpretability. Recent progress in large language model (LLM)-based agents has demonstrated the capacity to process multi-modal data and support complex investment decision-making. Building on these advances, we present \textbf{“MountainLion”}, a multi-modal, multi-agent system for financial trading that coordinates specialized LLM-based agents to interpret financial data and generate investment strategies. MountainLion processes textual news, candlestick charts, and trading signal charts to produce high-quality financial reports, while also enabling modification of reports and investment recommendations through data-driven user interaction and question answering. A central reflection module analyzes historical trading signals and outcomes to continuously refine decision processes, and the system is capable of real-time report analysis, summarization, and dynamic adjustment of investment strategies. Empirical results confirm that MountainLion systematically enriches technical price triggers with contextual macroeconomic and capital flow signals, providing a more interpretable, robust, and actionable investment framework that improves returns and strengthens investor confidence. ...

NMIXX: Domain-Adapted Neural Embeddings for Cross-Lingual eXploration of Finance ArXiv ID: 2507.09601 “View on arXiv” Authors: Hanwool Lee, Sara Yu, Yewon Hwang, Jonghyun Choi, Heejae Ahn, Sungbum Jung, Youngjae Yu Abstract General-purpose sentence embedding models often struggle to capture specialized financial semantics, especially in low-resource languages like Korean, due to domain-specific jargon, temporal meaning shifts, and misaligned bilingual vocabularies. To address these gaps, we introduce NMIXX (Neural eMbeddings for Cross-lingual eXploration of Finance), a suite of cross-lingual embedding models fine-tuned with 18.8K high-confidence triplets that pair in-domain paraphrases, hard negatives derived from a semantic-shift typology, and exact Korean-English translations. Concurrently, we release KorFinSTS, a 1,921-pair Korean financial STS benchmark spanning news, disclosures, research reports, and regulations, designed to expose nuances that general benchmarks miss. When evaluated against seven open-license baselines, NMIXX’s multilingual bge-m3 variant achieves Spearman’s rho gains of +0.10 on English FinSTS and +0.22 on KorFinSTS, outperforming its pre-adaptation checkpoint and surpassing other models by the largest margin, while revealing a modest trade-off in general STS performance. Our analysis further shows that models with richer Korean token coverage adapt more effectively, underscoring the importance of tokenizer design in low-resource, cross-lingual settings. By making both models and the benchmark publicly available, we provide the community with robust tools for domain-adapted, multilingual representation learning in finance. ...

A Framework for Predictive Directional Trading Based on Volatility and Causal Inference ArXiv ID: 2507.09347 “View on arXiv” Authors: Ivan Letteri Abstract Purpose: This study introduces a novel framework for identifying and exploiting predictive lead-lag relationships in financial markets. We propose an integrated approach that combines advanced statistical methodologies with machine learning models to enhance the identification and exploitation of predictive relationships between equities. Methods: We employed a Gaussian Mixture Model (GMM) to cluster nine prominent stocks based on their mid-range historical volatility profiles over a three-year period. From the resulting clusters, we constructed a multi-stage causal inference pipeline, incorporating the Granger Causality Test (GCT), a customised Peter-Clark Momentary Conditional Independence (PCMCI) test, and Effective Transfer Entropy (ETE) to identify robust, predictive linkages. Subsequently, Dynamic Time Warping (DTW) and a K-Nearest Neighbours (KNN) classifier were utilised to determine the optimal time lag for trade execution. The resulting strategy was rigorously backtested. Results: The proposed volatility-based trading strategy, tested from 8 June 2023 to 12 August 2023, demonstrated substantial efficacy. The portfolio yielded a total return of 15.38%, significantly outperforming the 10.39% return of a comparative Buy-and-Hold strategy. Key performance metrics, including a Sharpe Ratio up to 2.17 and a win rate up to 100% for certain pairs, confirmed the strategy’s viability. Conclusion: This research contributes a systematic and robust methodology for identifying profitable trading opportunities derived from volatility-based causal relationships. The findings have significant implications for both academic research in financial modelling and the practical application of algorithmic trading, offering a structured approach to developing resilient, data-driven strategies. ...

Functionally Generated Portfolios Under Stochastic Transaction Costs: Theory and Empirical Evidence ArXiv ID: 2507.09196 “View on arXiv” Authors: Nader Karimi, Erfan Salavati Abstract Assuming frictionless trading, classical stochastic portfolio theory (SPT) provides relative arbitrage strategies. However, the costs associated with real-world execution are state-dependent, volatile, and under increasing stress during liquidity shocks. Using an Ito diffusion that may be connected with asset prices, we extend SPT to a continuous-time equity market with proportional, stochastic transaction costs. We derive closed-form lower bounds on cost-adjusted relative wealth for a large class of functionally generated portfolios; these bounds provide sufficient conditions for relative arbitrage to survive random costs. A limit-order-book cost proxy in conjunction with a Milstein scheme validates the theoretical order-of-magnitude estimates. Finally, we use intraday bid-ask spreads as a stand-in for cost volatility in a back-test of CRSP small-cap data (1994–2024). Despite experiencing larger declines during the 2008 and 2020 liquidity crises, diversity- and entropy-weighted portfolios continue to beat the value-weighted benchmark by 3.6 and 2.9 percentage points annually, respectively, after cost deduction. ...

Joint deep calibration of the 4-factor PDV model ArXiv ID: 2507.09412 “View on arXiv” Authors: Fabio Baschetti, Giacomo Bormetti, Pietro Rossi Abstract Joint calibration to SPX and VIX market data is a delicate task that requires sophisticated modeling and incurs significant computational costs. The latter is especially true when pricing of volatility derivatives hinges on nested Monte Carlo simulation. One such example is the 4-factor Markov Path-Dependent Volatility (PDV) model of Guyon and Lekeufack (2023). Nonetheless, its realism has earned it considerable attention in recent years. Gazzani and Guyon (2025) marked a relevant contribution by learning the VIX as a random variable, i.e., a measurable function of the model parameters and the Markovian factors. A neural network replaces the inner simulation and makes the joint calibration problem accessible. However, the minimization loop remains slow due to expensive outer simulation. The present paper overcomes this limitation by learning SPX implied volatilities, VIX futures, and VIX call option prices. The pricing functions reduce to simple matrix-vector products that can be evaluated on the fly, shrinking calibration times to just a few seconds. ...

Arbitrage on Decentralized Exchanges ArXiv ID: 2507.08302 “View on arXiv” Authors: Xue Dong He, Chen Yang, Yutian Zhou Abstract Decentralized exchanges (DEXs) are alternative venues to centralized exchanges (CEXs) for trading cryptocurrencies and have become increasingly popular. An arbitrage opportunity arises when the exchange rate of two cryptocurrencies in a DEX differs from that in a CEX. Arbitrageurs can then trade on the DEX and CEX to make a profit. Trading on the DEX incurs a gas fee, which determines the priority of the trade being executed. We study a gas-fee competition game between two arbitrageurs who maximize their expected profit from trading. We derive the unique symmetric mixed Nash equilibrium and find that (i) the arbitrageurs may choose not to trade when the arbitrage opportunity and liquidity is small; (ii) the probability of the arbitrageurs choosing a higher gas fee is lower; (iii) the arbitrageurs pay a higher gas fee and trade more when the arbitrage opportunity becomes larger and when liquidity becomes higher; (iv) the arbitrageurs’ expected profit could increase with arbitrage opportunity and liquidity. The above findings are consistent with our empirical study. ...