https://arxiv.org/api/ugo1Jk6TymxwvaVq0zgNOsGjhBo 2026-06-21T19:10:35Z 3022 165 15 http://arxiv.org/abs/2511.15214v2 2025-11-25T18:42:49Z

Economic behavior is shaped not only by quantitative information but also by the narratives through which such information is communicated and interpreted (Shiller, 2017). I show that narratives extracted from earnings calls significantly improve the prediction of both realized earnings and analyst expectations. To uncover the underlying mechanisms, I introduce a novel text-morphing methodology in which large language models generate counterfactual transcripts that systematically vary topical emphasis (the prevailing narrative) while holding quantitative content fixed. This framework allows me to precisely measure how analysts under- and over-react to specific narrative dimensions. The results reveal systematic biases: analysts over-react to sentiment (optimism) and under-react to narratives of risk and uncertainty. Overall, the analysis offers a granular perspective on the mechanisms of expectation formation through the competing narratives embedded in corporate communication.

2025-11-19T08:06:46Z Giuseppe Matera http://arxiv.org/abs/2206.15365v10 2025-11-19T14:48:22Z

The false discovery rate (FDR) measures the share of false positives in a set of statistical tests. I develop simple and intuitive bounds on the FDR in cross-sectional predictability publications. The simplest bound requires just a few lines of math and finds $\text{FDR} \le 25\%$ based on summary statistics in eight out of nine previous studies. A more refined bound finds $\text{FDR} \le 9\%$. The FDR is small because randomly selecting accounting ratios produces statistically significant predictability far more often than would occur if there were no predictability. The bounds also reconcile the disparate FDR estimates in the literature.

2022-06-30T15:36:31Z Andrew Y. Chen http://arxiv.org/abs/2511.15456v1 2025-11-19T14:15:23Z

As Decentralized Finance (DeFi) develops, understanding user intent behind DeFi transactions is crucial yet challenging due to complex smart contract interactions, multifaceted on-/off-chain factors, and opaque hex logs. Existing methods lack deep semantic insight. To address this, we propose the Transaction Intent Mining (TIM) framework. TIM leverages a DeFi intent taxonomy built on grounded theory and a multi-agent Large Language Model (LLM) system to robustly infer user intents. A Meta-Level Planner dynamically coordinates domain experts to decompose multiple perspective-specific intent analyses into solvable subtasks. Question Solvers handle the tasks with multi-modal on/off-chain data. While a Cognitive Evaluator mitigates LLM hallucinations and ensures verifiability. Experiments show that TIM significantly outperforms machine learning models, single LLMs, and single Agent baselines. We also analyze core challenges in intent inference. This work helps provide a more reliable understanding of user motivations in DeFi, offering context-aware explanations for complex blockchain activity.

2025-11-19T14:15:23Z Written in 2025 Q1 Qian'ang Mao Yuxuan Zhang Jiaman Chen Wenjun Zhou Jiaqi Yan http://arxiv.org/abs/2511.15364v1 2025-11-19T11:44:48Z

We show that while anonymization effectively obscures firm identity, it significantly reduces the power of textual understanding, thereby diminishing models' ability to extract meaningful economic signals from financial texts. This information loss is particularly severe when numerical and object entities are removed from texts and is amplified in texts characterized by high linguistic uncertainty and firm specificity. Importantly, in the setting of sentiment extraction from earnings call transcripts, we find that information loss induced by anonymization is more pervasive and severe than the effects of look-ahead bias, suggesting that the costs of anonymization may outweigh its benefits in certain financial applications.

2025-11-19T11:44:48Z Ke Wu Baozhong Yang Zhenkun Ying Dexin Zhou http://arxiv.org/abs/2511.15123v1 2025-11-19T04:57:19Z

Financial event studies, ubiquitous in finance research, typically use linear factor models with known factors to estimate abnormal returns and identify causal effects of information events. This paper demonstrates that when factor models are misspecified -- an almost certain reality -- traditional event study estimators produce inconsistent estimates of treatment effects. The bias is particularly severe during volatile periods, over long horizons, and when event timing correlates with market conditions. We derive precise conditions for identification and expressions for asymptotic bias. As an alternative, we propose synthetic control methods that construct replicating portfolios from control securities without imposing specific factor structures. Revisiting four empirical applications, we show that some established findings may reflect model misspecification rather than true treatment effects. While traditional methods remain reliable for short-horizon studies with random event timing, our results suggest caution when interpreting long-horizon or volatile-period event studies and highlight the importance of quasi-experimental designs when available.

2025-11-19T04:57:19Z Paul Goldsmith-Pinkham Tianshu Lyu http://arxiv.org/abs/2512.02029v1 2025-11-19T03:46:37Z

Crypto enthusiasts claim that buying and holding crypto assets yields high returns, often citing Bitcoin's past performance to promote other tokens and fuel fear of missing out. However, understanding the real risk-return trade-off and what factors affect future crypto returns is crucial as crypto becomes increasingly accessible to retail investors through major brokerages. We examine the HODL strategy through two independent analyses. First, we implement 480 million Monte Carlo simulations across 378 non-stablecoin crypto assets, net of trading fees and the opportunity cost of 1-month Treasury bills, and find strong evidence of survivorship bias and extreme downside concentration. At the 2-3 year horizon, the median excess return is -28.4 percent, the 1 percent conditional value at risk indicates that tail scenarios wipe out principal after all costs, and only the top quartile achieves very large gains, with a mean excess return of 1,326.7 percent. These results challenge the HODL narrative: across a broad set of assets, simple buy-and-hold loads extreme downside risk onto most investors, and the miracles mostly belong to the luckiest quarter. Second, using a Bayesian multi-horizon local projection framework, we find that endogenous predictors based on realized risk-return metrics have economically negligible and unstable effects, while macro-finance factors, especially the 24-week exponential moving average of the Fear and Greed Index, display persistent long-horizon impacts and high cross-basket stability. Where significant, a one-standard-deviation sentiment shock reduces forward top-quartile mean excess returns by 15-22 percentage points and median returns by 6-10 percentage points over 1-3 year horizons, suggesting that macro-sentiment conditions, rather than realized return histories, are the dominant indicators for future outcomes.

2025-11-19T03:46:37Z Weikang Zhang Alison Watts http://arxiv.org/abs/2410.12825v4 2025-11-18T19:38:41Z

Users engage with financial services companies through multiple channels, often interacting with mobile applications, web platforms, call centers, and physical locations to service their accounts. The resulting interactions are recorded at heterogeneous temporal resolutions across these domains. This multi-channel data can be combined and encoded to create a comprehensive representation of the customer's journey for accurate intent prediction. This demands sequential learning solutions. NMT transformers achieve state-of-the-art sequential representation learning by encoding context and decoding for the next best action to represent long-range dependencies. However, three major challenges exist while combining multi-domain sequences within an encoder-decoder transformers architecture for intent prediction applications: a) aligning sequences with different sampling rates b) learning temporal dynamics across multi-variate, multi-domain sequences c) combining dynamic and static sequences. We propose an encoder-decoder transformer model to address these challenges for contextual and sequential intent prediction in financial servicing applications. Our experiments show significant improvement over the existing tabular method.

2024-10-01T16:22:07Z Accepted at RecTemp @ RecSys 2024, 6 pages, 3 figures Dwipam Katariya Juan Manuel Origgi Yage Wang Thomas Caputo http://arxiv.org/abs/2511.13878v1 2025-11-17T19:54:56Z

The Hongkong and Shanghai Banking Co (HSBC) just survived a civil war intermitted by World War II. By the 1950s, it obviously needed to close all its branches in Mao's People's Republic of China, yet could somehow hold its Shanghai branch, which continued likely in the shadows, as non-state banking was illegalised and even simple land owners were executed merely for being labelled "capitalist". This Asia-focused bank --in spite of it all-- grew from these conditions into the behemoth it is today. Part of the growth was based on the economic boom in its core market Hong Kong, to which HSBC likely also contributed. To expand and diversify, HSBC continued the growth strategy that already started since its early days in the 1860s, this time just also inorganically: It acquired other banks, in most cases fully and in other regions. The most important acquisition was the takeover of the roughly equally-sized UK-based Midland Bank; for the following reasons: 1) It came just a year after the 1991 change of HSBC's headquarters and place of incorporation to London, so HSBC could smoothly integrate with Midland. This step also came with an additional listing of securities in London, providing HSBC funds. 2) These funds were used efficiently without much idling for the humongous acquisition. 3) The preceding decade of Margaret Thatcher's banking and finance deregulation in the UK created a beneficial environment for HSBC. 4) HSBC was proven right by the developments in Hong Kong, where the Communist Party of China illegally eroded democracy, the rule of law and civil liberties since 2020, despite promising to maintain these at least until 1 July 2047. A list of likely reasons for HSBC's prosperity and stability in face of the at times hostile environments is also provided, from which business lessons can be drawn.

2025-11-17T19:54:56Z CM presented this research project at Die Lange Nacht der Wissenschaften 2025 in Erlangen on 2025-10-25 Christopher Mantzaris Ajda Fošner http://arxiv.org/abs/2510.11677v2 2025-11-17T18:56:19Z

We introduce a family of chronologically consistent, instruction-tuned large language models to eliminate lookahead bias. Each model is trained only on data available before a clearly defined knowledge-cutoff date, ensuring strict temporal separation from any post-cutoff data. The resulting framework offers (i) a simple, conversational chat interface, (ii) fully open, fixed model weights that guarantee replicability, and (iii) a conservative lower bound on forecast accuracy, isolating the share of predictability that survives once training leakage is removed. Together, these features provide researchers with an easy-to-use generative AI tool useful for a wide range of prediction tasks that is free of lookahead bias.

2025-10-13T17:45:24Z Songrun He Linying Lv Asaf Manela Jimmy Wu http://arxiv.org/abs/2407.19858v7 2025-11-17T01:54:48Z

In quantitative finance, machine learning methods are essential for alpha generation. This study introduces a new approach that combines Hidden Markov Models (HMM) and neural networks, integrated with Black-Litterman portfolio optimization. During the COVID period (2019-2022), this dual-model approach achieved a 83% return with a Sharpe ratio of 0.77. It incorporates two risk models to enhance risk management, showing efficiency during volatile periods. The methodology was implemented on the QuantConnect platform, which was chosen for its robust framework and experimental reproducibility. The system, which predicts future price movements, includes a three-year warm-up to ensure proper algorithm function. It targets highly liquid, large-cap energy stocks to ensure stable and predictable performance while also considering broker payments. The dual-model alpha system utilizes log returns to select the optimal state based on the historical performance. It combines state predictions with neural network outputs, which are based on historical data, to generate trading signals. This study examined the architecture of the trading system, data pre-processing, training, and performance. The full code and backtesting data are available under the QuantConnect terms: https://github.com/tiagomonteiro0715/AI-Powered-Energy-Algorithmic-Trading-Integrating-Hidden-Markov-Models-with-Neural-Networks

2024-07-29T10:26:52Z 25 pages, 4 figures, 2 tables Tiago Monteiro http://arxiv.org/abs/2509.01110v2 2025-11-16T23:13:08Z

We present NoLBERT, a lightweight, timestamped foundational language model for empirical research -- particularly for forecasting in economics, finance, and the social sciences. By pretraining exclusively on text from 1976 to 1995, NoLBERT avoids both lookback and lookahead biases (information leakage) that can undermine econometric inference. It exceeds domain-specific baselines on NLP benchmarks while maintaining temporal consistency. Applied to patent texts, NoLBERT enables the construction of firm-level innovation networks and shows that gains in innovation centrality predict higher long-run profit growth.

2025-09-01T04:07:10Z NeurIPS 2025 (GenAI in Finance) Ali Kakhbod Peiyao Li http://arxiv.org/abs/2511.11416v1 2025-11-14T15:46:13Z

The efficiency of pension schemes in Kenya invites elevated interest owing to the increasing pension contribution amounts and the expectation that benefits paid out of these schemes would protect members from old age poverty. The study investigates the intervening effect of risk management on the relationship between corporate governance and the efficiency of pension schemes in Kenya. The study employs panel data consisting of 896 observations from 128 schemes in a sample period from 2015 to 2021. The study finds that risk management significantly mediates the relationship between employee representatives on the board of trustees, as a component of corporate governance, and the efficiency of pension schemes. Consequently, the mediation effect of risk management indicates that when employee representatives are involved in governance, the presence of strong risk management practices ensures that their contributions lead to improved efficiency. Risk management, therefore, serves as a critical safeguard that enables governance structures to function more effectively and contribute to the overall performance of the scheme.

2025-11-14T15:46:13Z 21 pages Journal of Financial Risk Management, 2025, 14(3), 304-324 Sylvester Willys Namagwa 10.4236/jfrm.2025.143017 http://arxiv.org/abs/2511.10999v1 2025-11-14T06:41:30Z

As life expectancy in Kenya increases, so does the need for efficient pension schemes that can secure a dignified retirement and protect members from old age poverty. Limited research, however, has explored the efficiency of these schemes under existing governance structures. This study addresses that gap by examining the combined effects of corporate governance, risk management, and industry regulation on pension scheme efficiency in Kenya. Using a quantitative design, we conducted a panel regression analysis on a seven-year secondary dataset of 128 Kenyan pension schemes, totaling 896 observations. Our results reveal significant insights That the presence of employee representatives on the board and effective risk management have a significant positive effect on efficiency. Conversely, independent board members exhibit a significant negative effect. Other factors, including top management representation, female board members, and industry regulation, showed no significant effect on efficiency in the joint model. These findings suggest that the impact of governance and risk management on efficiency is nuanced, with specific factors like employee representation playing a more prominent role. We propose that the electoral process for employee board members may introduce a Self Cleaning Mechanism that progressively enhances scheme efficiency. This mechanism offers a novel theoretical extension of Agency Theory, explaining the convergence of interests between elected trustees and scheme members.

2025-11-14T06:41:30Z 28 pages Journal of Financial Risk Management, 2025, 14(4), 428-455 Sylvester Willys Namagwa 10.4236/jfrm.2025.144023 http://arxiv.org/abs/2511.10715v1 2025-11-13T14:16:40Z

Europe's largest bank by assets as of 2025 started out in the 1860s in one of Europe's colonies: The Hongkong and Shanghai Banking Co (HSBC). Multiple wars forced Qing China and later the young Republic of China into a series of unequal treaties, one of which was the forced legalisation of the opium trade from parts of the British Empire into China, another was opening several cities, including Shanghai, for trade and granting extensive civil, property and business rights to non-residents and yet another was the annexation of Hong Kong by the United Kingdom. These are the conditions that created HSBC and in which it thrived, including from opium-related profits. During periods of relative calm, the bank grew geographically and made profits -- whether in moral or unethical, whether in legal or unlawful ways -- which helped HSBC weather the storms of civil and world wars. Other aspects contributed to HSBC's survival and success, such as its global nature, which allowed it to diversify and shift away from regions when danger emerges there and find shelter in safer havens. Yet the resilient survival abilities and the financial success of HSBC until 1950 should not distract from the fact that in addition to its tainted cradle and early profits from facilitating the poisoning of a whole society, its human resource system was also discrimination based, attempting to divide the one human race into different groups -- in spite of such practises being opposed to medical and biological facts. What is particularly interesting to see is that not only has HSBC yet to apologise for the early drug-related blood money it made: it even fails to mention its colonial, drug profits tainted past on any of its many history-themed pages sighted. Later sections contain possible reasons for HSBC's resilience and success, particularly interesting for entrepreneurs.

2025-11-13T14:16:40Z CM presented this research project at Die Lange Nacht der Wissenschaften 2025 in Erlangen on 2025-10-25 Christopher Mantzaris Ajda Fošner http://arxiv.org/abs/2511.05508v2 2025-11-13T09:54:07Z

Financial advisors and investors struggle with information overload from financial news, where irrelevant content and noise obscure key market signals and hinder timely investment decisions. To address this, we propose a novel Chain-of-Thought (CoT) summarization framework that condenses financial news into concise, event-driven summaries. The framework integrates user-specified keywords to generate personalized outputs, ensuring that only the most relevant contexts are highlighted. These personalized summaries provide an intermediate layer that supports language models in producing investor-focused narratives, bridging the gap between raw news and actionable insights.

2025-10-24T05:55:05Z Proceedings of ICDM Workshops Tianyi Zhang Mu Chen