|
on Financial Markets |
| By: | Zefeng Chen; Darcy Pu |
| Abstract: | Can fully agentic AI nowcast stock returns? We deploy a state-of-the-art Large Language Model to evaluate the attractiveness of each Russell 1000 stock daily, starting from April 2025 when AI web interfaces enabled real-time search. Our data contribution is unique along three dimensions. First, the nowcasting framework is completely out-of-sample and free of look-ahead bias by construction: predictions are collected at the current edge of time, ensuring the AI has no knowledge of future outcomes. Second, this temporal design is irreproducible -- once the information environment passes, it can never be recreated. Third, our framework is 100% agentic: we do not feed the model news, disclosures, or curated text; it autonomously searches the web, filters sources, and synthesises information into quantitative predictions. We find that AI possesses genuine stock selection ability, but only for identifying top winners. Longing the 20 highest-ranked stocks generates a daily Fama-French five-factor plus momentum alpha of 18.4 basis points and an annualised Sharpe ratio of 2.43. Critically, these returns derive from an implementable strategy trading highly liquid Russell 1000 constituents, with transaction costs representing less than 10\% of gross alpha. However, this predictability is highly concentrated: expanding beyond the top tier rapidly dilutes alpha, and bottom-ranked stocks exhibit returns statistically indistinguishable from the market. We hypothesise that this asymmetry reflects online information structure: genuinely positive news generates coherent signals, while negative news is contaminated by strategic corporate obfuscation and social media noise. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2601.11958 |
| By: | Maxwell Block |
| Abstract: | This study examines short-term market responses to material cybersecurity incidents disclosed under Item 1.05 of Form 8-K. Drawing on a sample of disclosures made between 2023 and 2025, daily stock price movements were evaluated over a standardized event window surrounding each filing. On average, companies experienced negative price reactions following the disclosure of a material cybersecurity incident. Comparisons across company characteristics indicate that smaller companies tended to incur more pronounced declines, while differences by sector and beta were not evident. These findings offer empirical insight into how public markets interpret cybersecurity risks when they are formally reported and suggest that firm size may influence the degree of sensitivity to such events. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.06144 |
| By: | Bohan Liang; Zijian Chen; Qi Jia; Kaiwei Zhang; Kaiyuan Ji; Guangtao Zhai |
| Abstract: | Stock prediction, a subject closely related to people's investment activities in fully dynamic and live environments, has been widely studied. Current large language models (LLMs) have shown remarkable potential in various domains, exhibiting expert-level performance through advanced reasoning and contextual understanding. In this paper, we introduce PriceSeer, a live, dynamic, and data-uncontaminated benchmark specifically designed for LLMs performing stock prediction tasks. Specifically, PriceSeer includes 110 U.S. stocks from 11 industrial sectors, with each containing 249 historical data points. Our benchmark implements both internal and external information expansion, where LLMs receive extra financial indicators, news, and fake news to perform stock price prediction. We evaluate six cutting-edge LLMs under different prediction horizons, demonstrating their potential in generating investment strategies after obtaining accurate price predictions for different sectors. Additionally, we provide analyses of LLMs' suboptimal performance in long-term predictions, including the vulnerability to fake news and specific industries. The code and evaluation data will be open-sourced at https://github.com/BobLiang2113/PriceSee r. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2601.06088 |
| By: | Varun Narayan Kannan Pillai; Akshay Ajith; Sumesh K J |
| Abstract: | The intricate behavior patterns of financial markets are influenced by fundamental, technical, and psychological factors. During times of high volatility and regime shifts causes many traditional strategies like trend-following or mean-reversion to fail. This paper proposes a hybrid AI-based trading strategy that combines (1) trend-following and directional momentum capture via EMA and MACD, (2) detection of price normalization through mean-reversion using RSI and Bollinger Bands, (3) market psychological interpretation through sentiment analysis using FinBERT, (4) signal generation through machine learning using XGBoost and (5)dynamically adjusting exposure with market regime filtering based on volatility and return environments. The system achieved a final portfolio value of $235, 492.83, yielding a return of 135.49% on initial investment over a period of 24 months. The hybrid model outperformed major benchmark indexes like S&P 500 and NASDAQ-100 over the same period showing strong flexibility and lower downside risk with superior profits validating the use of multi-modal AI in algorithmic trading. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2601.19504 |
| By: | Matthew Brigida |
| Abstract: | We estimate risk premia in the cross-section of cryptocurrency returns using the Giglio-Xiu (2021) three-pass approach, allowing for omitted latent factors alongside observed stock-market and crypto-market factors. Using weekly data on a broad universe of large cryptocurrencies, we find that crypto expected returns load on both crypto-specific factors and selected equity-industry factors associated with technology and profitability, consistent with increased integration between crypto and traditional markets. In addition, we study non-tradable state variables capturing investor sentiment (Fear and Greed), speculative rotation (Altcoin Season Index), and security shocks (hacked value scaled by market capitalization), which are new to the literature. Relative to conventional Fama-MacBeth estimates, the latent-factor approach yields materially different premia for key factors, highlighting the importance of controlling for unobserved risks in crypto asset pricing. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2601.07664 |
| By: | Habib Badawi; Mohamed Hani; Taufikin Taufikin |
| Abstract: | Financial markets often appear chaotic, yet ranges are rarely accidental. They emerge from structured interactions between market context and capital conditions. The four-hour timeframe provides a critical lens for observing this equilibrium zone where institutional positioning, leveraged exposure, and liquidity management converge. Funding mechanisms, especially in perpetual futures, act as disciplinary forces that regulate trader behavior, impose economic costs, and shape directional commitment. When funding aligns with the prevailing 4H context, price expansion becomes possible; when it diverges, compression and range-bound behavior dominate. Ranges therefore represent controlled balance rather than indecision, reflecting strategic positioning by informed participants. Understanding how 4H context and funding operate as market governors is essential for interpreting cryptocurrency price action as a rational, power-mediated process. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2601.06084 |
| By: | Janda, Karel; Rozsahegyi, Marketa; Quang Van Tran; Zhang, Binyi |
| Abstract: | This paper investigates the role of investor sentiment in the pricing and volatility dynamics of green bond exchange-traded funds (ETFs). For a construction of green sentiment one original and two already existing natural language processing models are used and evaluated. The VAR model found no significant impact of green sentiment on ETF returns. The GARCH (1, 1) estimation strongly supported the presence of volatility clustering and time-varying volatility in green bond ETF returns, validating the use of conditional heteroskedasticity models. Regressing the conditional volatility on sentiment scores revealed a significant negative relationship – higher sentiment is associated with lower volatility. This finding implies that positive green sentiment contributes to market stability and may reduce perceived risk, reinforcing the importance of investor psychology in green financial markets. |
| Keywords: | Machine learning, NLP model, ESG, Exchange Traded Funds |
| JEL: | C45 C55 G11 G17 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:esprep:335550 |
| By: | Hamoon Soleimani |
| Abstract: | Since its inception, Bitcoin has been positioned as a revolutionary alternative to national currencies, attracting immense public and academic interest. This paper presents a critical evaluation of this claim, suggesting that Bitcoin faces significant structural barriers to qualifying as money. It synthesizes critiques from two distinct schools of economic thought - Post-Keynesianism and the Austrian School - and validates their conclusions with rigorous technical analysis. From a Post-Keynesian perspective, it is argued that Bitcoin does not function as money because it is not a debt-based IOU and fails to exhibit the essential properties required for a stable monetary asset (Vianna, 2021). Concurrently, from an Austrian viewpoint, it is shown to be inconsistent with a strict interpretation of Mises's Regression Theorem, as it lacks prior non-monetary value and has not achieved the status of the most saleable commodity (Peniaz and Kavaliou, 2024). These theoretical arguments are then supported by an empirical analysis of Bitcoin's extreme volatility, hard-coded scalability limits, fragile market structure, and insecure long-term economic design. The paper concludes that Bitcoin is more accurately characterized as a novel speculative asset whose primary legacy may be the technological innovation it has spurred, rather than its viability as a monetary standard. |
| Date: | 2025–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.07840 |
| By: | Timo Dimitriadis; Yannick Hoga |
| Abstract: | Following several episodes of financial market turmoil in recent decades, changes in systemic risk have drawn growing attention. Therefore, we propose surveillance schemes for systemic risk, which allow to detect misspecified systemic risk forecasts in an "online" fashion. This enables daily monitoring of the forecasts while controlling for the accumulation of false test rejections. Such online schemes are vital in taking timely countermeasures to avoid financial distress. Our monitoring procedures allow multiple series at once to be monitored, thus increasing the likelihood and the speed at which early signs of trouble may be picked up. The tests hold size by construction, such that the null of correct systemic risk assessments is only rejected during the monitoring period with (at most) a pre-specified probability. Monte Carlo simulations illustrate the good finite-sample properties of our procedures. An empirical application to US banks during multiple crises demonstrates the usefulness of our surveillance schemes for both regulators and financial institutions. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2601.08598 |
| By: | Junwei Yang |
| Abstract: | This study explores contagion in the Chinese stock market using Hawkes processes to analyze autocorrelation and cross-correlation in multivariate time series data. We examine whether market indices exhibit trending behavior and whether sector indices influence one another. By fitting self-exciting and inhibitory Hawkes processes to daily returns of indices like the Shanghai Composite, Shenzhen Component, and ChiNext, as well as sector indices (CSI Consumer, Healthcare, and Financial), we identify long- term dependencies and trending patterns, including upward, downward, and over- sold rebound trends. Results show that during high trading activity, sector indices tend to sustain their trends, while low activity periods exhibit strong sector rotation. This research models stock price movements using spatiotemporal Hawkes processes, leveraging conditional intensity functions to explain sector rotation, advancing the understanding of financial contagion. |
| Date: | 2025–12 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2512.08000 |