Equities

Realized Volatility Forecasting for New Issues and Spin-Offs using Multi-Source Transfer Learning ArXiv ID: 2503.12648 “View on arXiv” Authors: Unknown Abstract Forecasting the volatility of financial assets is essential for various financial applications. This paper addresses the challenging task of forecasting the volatility of financial assets with limited historical data, such as new issues or spin-offs, by proposing a multi-source transfer learning approach. Specifically, we exploit complementary source data of assets with a substantial historical data record by selecting source time series instances that are most similar to the limited target data of the new issue/spin-off. Based on these instances and the target data, we estimate linear and non-linear realized volatility models and compare their forecasting performance to forecasts of models trained exclusively on the target data, and models trained on the entire source and target data. The results show that our transfer learning approach outperforms the alternative models and that the integration of complementary data is also beneficial immediately after the initial trading day of the new issue/spin-off. ...

A Deep Reinforcement Learning Approach to Automated Stock Trading, using xLSTM Networks ArXiv ID: 2503.09655 “View on arXiv” Authors: Unknown Abstract Traditional Long Short-Term Memory (LSTM) networks are effective for handling sequential data but have limitations such as gradient vanishing and difficulty in capturing long-term dependencies, which can impact their performance in dynamic and risky environments like stock trading. To address these limitations, this study explores the usage of the newly introduced Extended Long Short Term Memory (xLSTM) network in combination with a deep reinforcement learning (DRL) approach for automated stock trading. Our proposed method utilizes xLSTM networks in both actor and critic components, enabling effective handling of time series data and dynamic market environments. Proximal Policy Optimization (PPO), with its ability to balance exploration and exploitation, is employed to optimize the trading strategy. Experiments were conducted using financial data from major tech companies over a comprehensive timeline, demonstrating that the xLSTM-based model outperforms LSTM-based methods in key trading evaluation metrics, including cumulative return, average profitability per trade, maximum earning rate, maximum pullback, and Sharpe ratio. These findings mark the potential of xLSTM for enhancing DRL-based stock trading systems. ...

Agent Trading Arena: A Study on Numerical Understanding in LLM-Based Agents ArXiv ID: 2502.17967 “View on arXiv” Authors: Unknown Abstract Large language models (LLMs) have demonstrated remarkable capabilities in natural language tasks, yet their performance in dynamic, real-world financial environments remains underexplored. Existing approaches are limited to historical backtesting, where trading actions cannot influence market prices and agents train only on static data. To address this limitation, we present the Agent Trading Arena, a virtual zero-sum stock market in which LLM-based agents engage in competitive multi-agent trading and directly impact price dynamics. By simulating realistic bid-ask interactions, our platform enables training in scenarios that closely mirror live markets, thereby narrowing the gap between training and evaluation. Experiments reveal that LLMs struggle with numerical reasoning when given plain-text data, often overfitting to local patterns and recent values. In contrast, chart-based visualizations significantly enhance both numerical reasoning and trading performance. Furthermore, incorporating a reflection module yields additional improvements, especially with visual inputs. Evaluations on NASDAQ and CSI datasets demonstrate the superiority of our method, particularly under high volatility. All code and data are available at https://github.com/wekjsdvnm/Agent-Trading-Arena. ...

Why do financial prices exhibit Brownian motion despite predictable order flow? ArXiv ID: 2502.17906 “View on arXiv” Authors: Unknown Abstract In financial market microstructure, there are two enigmatic empirical laws: (i) the market-order flow has predictable persistence due to metaorder splitters by institutional investors, well formulated as the Lillo-Mike-Farmer model. However, this phenomenon seems paradoxical given the diffusive and unpredictable price dynamics; (ii) the price impact $I(Q)$ of a large metaorder $Q$ follows the square-root law, $I(Q)\propto \sqrt{“Q”}$. Here we theoretically reveal why price dynamics follows Brownian motion despite predictable order flow by unifying these enigmas. We generalize the Lillo-Mike-Farmer model to nonlinear price-impact dynamics, which is mapped to an exactly solvable Lévy-walk model. Our exact solution shows that the price dynamics remains diffusive under the square-root law, even under persistent order flow. This work illustrates the crucial role of the square-root law in mitigating large price movements by large metaorders, thereby leading to the Brownian price dynamics, consistently with the efficient market hypothesis over long timescales. ...

Event-Based Limit Order Book Simulation under a Neural Hawkes Process: Application in Market-Making ArXiv ID: 2502.17417 “View on arXiv” Authors: Unknown Abstract In this paper, we propose an event-driven Limit Order Book (LOB) model that captures twelve of the most observed LOB events in exchange-based financial markets. To model these events, we propose using the state-of-the-art Neural Hawkes process, a more robust alternative to traditional Hawkes process models. More specifically, this model captures the dynamic relationships between different event types, particularly their long- and short-term interactions, using a Long Short-Term Memory neural network. Using this framework, we construct a midprice process that captures the event-driven behavior of the LOB by simulating high-frequency dynamics like how they appear in real financial markets. The empirical results show that our model captures many of the broader characteristics of the price fluctuations, particularly in terms of their overall volatility. We apply this LOB simulation model within a Deep Reinforcement Learning Market-Making framework, where the trading agent can now complete trade order fills in a manner that closely resembles real-market trade execution. Here, we also compare the results of the simulated model with those from real data, highlighting how the overall performance and the distribution of trade order fills closely align with the same analysis on real data. ...

Ensemble RL through Classifier Models: Enhancing Risk-Return Trade-offs in Trading Strategies ArXiv ID: 2502.17518 “View on arXiv” Authors: Unknown Abstract This paper presents a comprehensive study on the use of ensemble Reinforcement Learning (RL) models in financial trading strategies, leveraging classifier models to enhance performance. By combining RL algorithms such as A2C, PPO, and SAC with traditional classifiers like Support Vector Machines (SVM), Decision Trees, and Logistic Regression, we investigate how different classifier groups can be integrated to improve risk-return trade-offs. The study evaluates the effectiveness of various ensemble methods, comparing them with individual RL models across key financial metrics, including Cumulative Returns, Sharpe Ratios (SR), Calmar Ratios, and Maximum Drawdown (MDD). Our results demonstrate that ensemble methods consistently outperform base models in terms of risk-adjusted returns, providing better management of drawdowns and overall stability. However, we identify the sensitivity of ensemble performance to the choice of variance threshold τ, highlighting the importance of dynamic τ adjustment to achieve optimal performance. This study emphasizes the value of combining RL with classifiers for adaptive decision-making, with implications for financial trading, robotics, and other dynamic environments. ...

Analyzing Communicability and Connectivity in the Indian Stock Market During Crises ArXiv ID: 2502.08242 “View on arXiv” Authors: Unknown Abstract Understanding how information flows through the financial networks is important, especially during times of market turbulence. Unlike traditional assumptions where information travels along the shortest paths, real-world diffusion processes often follow multiple routes. To capture this complexity, we apply communicability, a network measure that quantifies the ease of information flow between nodes, even beyond the shortest path. In this study, we aim to examine how communicability responds to structural disruptions in financial networks during periods of high volatility. We compute communicability-based metrics on correlation-derived networks constructed from financial market data, and apply statistical testing through permutation methods to identify significant shifts in network structure. Our results show that approximately 70% and 80% of stock pairs exhibit statistically significant changes in communicability during the global financial crisis and the unprecedented COVID-19 crisis, respectively, at a significance level of 0.001. The observed shifts in shortest communicability path lengths offer directional cues about the nature and depth of each crisis. Furthermore, when used as features in machine learning classification models, communicability measures outperform the shortest-path-based measures in distinguishing between market stability and volatility periods. The performance of geometric measures was also comparable to that of topology-based measures. These findings offer valuable insights into the dynamic behavior of financial markets during times of crises and underscore the practical relevance of communicability in modeling systemic risk and information diffusion in complex networks. ...

Trend-encoded Probabilistic Multi-order Model: A Non-Machine Learning Approach for Enhanced Stock Market Forecasts ArXiv ID: 2502.08144 “View on arXiv” Authors: Unknown Abstract In recent years, the dominance of machine learning in stock market forecasting has been evident. While these models have shown decreasing prediction errors, their robustness across different datasets has been a concern. A successful stock market prediction model minimizes prediction errors and showcases robustness across various data sets, indicating superior forecasting performance. This study introduces a novel multiple lag order probabilistic model based on trend encoding (TeMoP) that enhances stock market predictions through a probabilistic approach. Results across different stock indexes from nine countries demonstrate that the TeMoP outperforms the state-of-the-art machine learning models in predicting accuracy and stabilization. ...

FinRL-DeepSeek: LLM-Infused Risk-Sensitive Reinforcement Learning for Trading Agents ArXiv ID: 2502.07393 “View on arXiv” Authors: Unknown Abstract This paper presents a novel risk-sensitive trading agent combining reinforcement learning and large language models (LLMs). We extend the Conditional Value-at-Risk Proximal Policy Optimization (CPPO) algorithm, by adding risk assessment and trading recommendation signals generated by a LLM from financial news. Our approach is backtested on the Nasdaq-100 index benchmark, using financial news data from the FNSPID dataset and the DeepSeek V3, Qwen 2.5 and Llama 3.3 language models. The code, data, and trading agents are available at: https://github.com/benstaf/FinRL_DeepSeek ...

Intraday order transition dynamics in high, medium, and low market cap stocks: A Markov chain approach ArXiv ID: 2502.07625 “View on arXiv” Authors: Unknown Abstract An empirical stochastic analysis of high-frequency, tick-by-tick order data of NASDAQ100 listed stocks is conducted using a first-order discrete-time Markov chain model to explore intraday order transition dynamics. This analysis focuses on three market cap categories: High, Medium, and Low. Time-homogeneous transition probability matrices are estimated and compared across time-zones and market cap categories, and we found that limit orders exhibit higher degree of inertia (DoI), i.e., the probability of placing consecutive limit order is higher, during the opening hour. However, in the subsequent hour, the DoI of limit order decreases, while that of market order increases. Limit order adjustments via additions and deletions of limit orders increases significantly after the opening hour. All the order transitions then stabilize during mid-hours. As the closing hour approaches, consecutive order executions surge, with decreased placement of buy and sell limit orders following sell and buy executions, respectively. In terms of the differences in order transitions between stocks of different market cap, DoI of orders is stronger in high and medium market cap stocks. On the other hand, lower market cap stocks show a higher probability of limit order modifications and greater likelihood of submitting sell/buy limit orders after buy/sell executions. Further, order transitions are clustered across all stocks, except during opening and closing hours. The findings of this study may be useful in understanding intraday order placement dynamics across stocks of varying market cap, thus aiding market participants in making informed order placements at different times of trading hour. ...