定量金融

• 新提交
• 交叉列表
• 替换

查看 最近的 文章

显示 2025年10月20日， 星期一 新的列表

新提交 (展示 12 之 12 条目 )

[1] arXiv:2510.14985 [中文pdf, pdf, 其他]

标题： DeepAries：自适应再平衡区间选择以增强投资组合选择 显示英文标题

标题： DeepAries: Adaptive Rebalancing Interval Selection for Enhanced Portfolio Selection

Jinkyu Kim, Hyunjung Yi, Mogan Gim, Donghee Choi, Jaewoo Kang

评论： CIKM 2025 应用研究赛道接受

主题： 投资组合管理 (q-fin.PM) ; 人工智能 (cs.AI) ; 计算工程、金融与科学 (cs.CE)

我们提出 DeepAries，一种用于动态投资组合管理的新颖深度强化学习框架，该框架联合优化再平衡决策的时间和分配。 与之前不考虑市场状况的固定再平衡间隔的强化学习方法不同，DeepAries 适应性地选择最优再平衡间隔以及投资组合权重，以减少不必要的交易成本并最大化风险调整后的收益。 我们的框架结合基于 Transformer 的状态编码器，有效捕捉复杂的长期市场依赖关系，并与近端策略优化（PPO）相结合，生成同时的离散（再平衡间隔）和连续（资产分配）动作。 在多个现实金融市场上的大量实验表明，DeepAries 在风险调整收益、交易成本和回撤方面显著优于传统固定频率和完全再平衡策略。 此外，我们在 https://deep-aries.github.io/ 提供 DeepAries 的实时演示，同时在 https://github.com/dmis-lab/DeepAries 上提供源代码和数据集，展示了 DeepAries 产生与市场制度变化一致的可解释再平衡和分配决策的能力。 总体而言，DeepAries 通过将时间安排和分配整合到统一的决策过程中，为自适应且实用的投资组合管理引入了一种创新范式。 显示英文摘要

We propose DeepAries , a novel deep reinforcement learning framework for dynamic portfolio management that jointly optimizes the timing and allocation of rebalancing decisions. Unlike prior reinforcement learning methods that employ fixed rebalancing intervals regardless of market conditions, DeepAries adaptively selects optimal rebalancing intervals along with portfolio weights to reduce unnecessary transaction costs and maximize risk-adjusted returns. Our framework integrates a Transformer-based state encoder, which effectively captures complex long-term market dependencies, with Proximal Policy Optimization (PPO) to generate simultaneous discrete (rebalancing intervals) and continuous (asset allocations) actions. Extensive experiments on multiple real-world financial markets demonstrate that DeepAries significantly outperforms traditional fixed-frequency and full-rebalancing strategies in terms of risk-adjusted returns, transaction costs, and drawdowns. Additionally, we provide a live demo of DeepAries at https://deep-aries.github.io/, along with the source code and dataset at https://github.com/dmis-lab/DeepAries, illustrating DeepAries' capability to produce interpretable rebalancing and allocation decisions aligned with shifting market regimes. Overall, DeepAries introduces an innovative paradigm for adaptive and practical portfolio management by integrating both timing and allocation into a unified decision-making process.

[2] arXiv:2510.14986 [中文pdf, pdf, html, 其他]

标题： RegimeFolio：一种用于动态市场中行业组合优化的制度感知机器学习系统 显示英文标题

标题： RegimeFolio: A Regime Aware ML System for Sectoral Portfolio Optimization in Dynamic Markets

Yiyao Zhang, Diksha Goel, Hussain Ahmad, Claudia Szabo

主题： 投资组合管理 (q-fin.PM) ; 人工智能 (cs.AI)

金融市场本质上是非平稳的，波动率制度的转变会改变资产共同运动和收益分布。 传统的投资组合优化方法通常基于平稳性或制度无关的假设，在适应这些变化时存在困难。 为了解决这些问题，我们提出了RegimeFolio，这是一种新的制度感知且行业专业化的框架。与现有的制度无关模型如DeepVol和DRL优化器不同，该框架将显式的波动率制度分割与行业特定的集成预测和自适应均值-方差配置相结合。 这种模块化架构确保预测和投资组合决策与当前市场状况保持一致，从而在动态市场中提高稳健性和可解释性。 RegimeFolio包含三个组件：(i) 一种基于VIX的可解释分类器用于市场制度检测；(ii) 针对制度和行业的集成学习者（随机森林、梯度提升）以捕捉条件收益结构；以及(iii) 一种具有收缩正则化协方差估计的动态均值-方差优化器，用于制度感知的配置。 我们在2020年至2024年间的34只大型美国股票上评估了RegimeFolio。该框架实现了137%的累计收益，夏普比率1.17，最大回撤降低了12%，并且在预测准确性方面比传统和先进的机器学习基准提高了15%到20%。 这些结果表明，在预测学习和投资组合配置中显式建模波动率制度可以提高稳健性，并在实际市场中实现更可靠的决策。 显示英文摘要

Financial markets are inherently non-stationary, with shifting volatility regimes that alter asset co-movements and return distributions. Standard portfolio optimization methods, typically built on stationarity or regime-agnostic assumptions, struggle to adapt to such changes. To address these challenges, we propose RegimeFolio, a novel regime-aware and sector-specialized framework that, unlike existing regime-agnostic models such as DeepVol and DRL optimizers, integrates explicit volatility regime segmentation with sector-specific ensemble forecasting and adaptive mean-variance allocation. This modular architecture ensures forecasts and portfolio decisions remain aligned with current market conditions, enhancing robustness and interpretability in dynamic markets. RegimeFolio combines three components: (i) an interpretable VIX-based classifier for market regime detection; (ii) regime and sector-specific ensemble learners (Random Forest, Gradient Boosting) to capture conditional return structures; and (iii) a dynamic mean-variance optimizer with shrinkage-regularized covariance estimates for regime-aware allocation. We evaluate RegimeFolio on 34 large cap U.S. equities from 2020 to 2024. The framework achieves a cumulative return of 137 percent, a Sharpe ratio of 1.17, a 12 percent lower maximum drawdown, and a 15 to 20 percent improvement in forecast accuracy compared to conventional and advanced machine learning benchmarks. These results show that explicitly modeling volatility regimes in predictive learning and portfolio allocation enhances robustness and leads to more dependable decision-making in real markets.

[3] arXiv:2510.14988 [中文pdf, pdf, html, 其他]

标题： 选择置信区间用于等权重投资组合 显示英文标题

标题： Selection Confidence Sets for Equally Weighted Portfolios

Davide Ferrari, Alessandro Fulci, Sandra Paterlini

主题： 投资组合管理 (q-fin.PM) ; 方法论 (stat.ME)

在N个资产的宇宙中，投资者通常通过选择资产子集来形成等权重组合（EWPs）。 EWPs简单、稳健且在样本外具有竞争力，但关于哪个子集真正表现最好的不确定性却被很大程度上忽略了。 传统方法通常依赖于一个选定的组合，但这未能考虑在考虑统计不确定性时可能表现同样良好的其他投资策略。 为解决这种选择不确定性，我们引入了等权重组合的选择置信集（SCS）：在给定损失函数和指定置信水平下，包含重复抽样下未知最优组合的组合集合。 SCS通过识别一组合理的组合来量化选择不确定性，挑战了唯一最优选择的概念。 像置信集一样，其大小反映了不确定性——在数据嘈杂或有限时增大，在样本量增加时减小。 理论上，我们证明了SCS以高概率覆盖未知的最优选择，并描述了其大小如何随着基本不确定性增长，并通过蒙特卡洛实验验证了这些结果。 对法国17个行业组合和第一层加密货币的应用强调了在比较等权重策略时考虑选择不确定性的必要性。 显示英文摘要

Given a universe of N assets, investors often form equally weighted portfolios (EWPs) by selecting subsets of assets. EWPs are simple, robust, and competitive out-of-sample, yet the uncertainty about which subset truly performs best is largely ignored. Traditional approaches typically rely on a single selected portfolio, but this fails to consider alternative investment strategies that may perform just as well when accounting for statistical uncertainty. To address this selection uncertainty, we introduce the Selection Confidence Set (SCS) for EWPs: the set of all portfolios that, under a given loss function and at a specified confidence level, contains the unknown set of optimal portfolios under repeated sampling. The SCS quantifies selection uncertainty by identifying a range of plausible portfolios, challenging the idea of a uniquely optimal choice. Like a confidence set, its size reflects uncertainty -- growing with noisy or limited data, and shrinking as the sample size increases. Theoretically, we establish that the SCS covers the unknown optimal selection with high probability and characterize how its size grows with underlying uncertainty, corroborating these results through Monte Carlo experiments. Applications to the French 17-Industry Portfolios and Layer-1 cryptocurrencies underscore the importance of accounting for selection uncertainty when comparing equally weighted strategies.

[4] arXiv:2510.15121 [中文pdf, pdf, 其他]

标题： 用于美国制造业供应链材料效率评估的物理扩展投入产出框架 显示英文标题

标题： A physically extended EEIO framework for material efficiency assessment in United States manufacturing supply chains

Heather Liddell, Beth Kelley, Liz Wachs, Alberta Carpenter, Joe Cresko

评论： 9页，4图。接受的手稿，将在2025年REMADE循环经济会议和技术峰会（华盛顿特区，2025年4月10-11日）上发表

主题： 一般经济学 (econ.GN) ; 计算机与社会 (cs.CY)

对经济中物质流的物理评估（例如，物质流量化）可以通过提供工业生产数量和供应链关系的有形物理背景，支持可持续脱碳和循环经济策略的发展。 然而，由于高质量原始数据的稀缺以及在数据报告中使用的行业分类系统之间缺乏统一性，完成物理评估具有挑战性。 本文描述了美国能源部（DOE）工业脱碳的能源环境投入产出（EEIO-IDA）模型的新物理扩展，生成了一个既在物理上又在环境上扩展的EEIO模型。 在模型框架中，美国经济被划分为商品生产和服务业子部门，每个商品生产子部门的质量流通过贸易数据（例如，联合国贸易数据库）和物理生产数据（例如，美国地质调查局）的组合进行量化。 鉴于并非所有子部门都有原始生产数据，开发并使用了价格插补和质量平衡假设，以高质量估算完成物理流数据集。 当该结果数据集与EEIO-IDA工具集成时，可以针对每个工业子部门在质量基础上量化环境影响强度指标（例如，CO$_2$eq/kg）。 这项工作旨在与现有的DOE框架和工具保持一致，包括EEIO-IDA工具、DOE工业脱碳路线图（2022年）以及美国工业转型路径研究（2025年）。 显示英文摘要

A physical assessment of material flows in an economy (e.g., material flow quantification) can support the development of sustainable decarbonization and circularity strategies by providing the tangible physical context of industrial production quantities and supply chain relationships. However, completing a physical assessment is challenging due to the scarcity of high-quality raw data and poor harmonization across industry classification systems used in data reporting. Here we describe a new physical extension for the U.S. Department of Energy's (DOE's) EEIO for Industrial Decarbonization (EEIO-IDA) model, yielding an expanded EEIO model that is both physically and environmentally extended. In the model framework, the U.S. economy is divided into goods-producing and service-producing subsectors, and mass flows are quantified for each goods-producing subsector using a combination of trade data (e.g., UN Comtrade) and physical production data (e.g., U.S. Geological Survey). Given that primary-source production data are not available for all subsectors, price-imputation and mass-balance assumptions are developed and used to complete the physical flows dataset with high-quality estimations. The resulting dataset, when integrated with the EEIO-IDA tool, enables the quantification of environmental impact intensity metrics on a mass basis (e.g., CO$_2$eq/kg)) for each industrial subsector. This work is designed to align with existing DOE frameworks and tools, including the EEIO-IDA tool, the DOE Industrial Decarbonization Roadmap (2022), and Pathways for U.S. Industrial Transformations study (2025).

[5] arXiv:2510.15288 [中文pdf, pdf, html, 其他]

标题： 印尼银行业股票的稳健优化组合优化 显示英文标题

标题： Portfolio Optimization of Indonesian Banking Stocks Using Robust Optimization

Visca Tri Winarty, Sena Safarina

主题： 投资组合管理 (q-fin.PM) ; 优化与控制 (math.OC)

自新冠疫情爆发以来，印度尼西亚证券交易所的投资者数量持续增加，强调了投资组合优化在平衡风险和收益中的重要性。 经典均值-方差优化模型虽然被广泛使用，但依赖于不确定的历史收益率和风险估计，可能导致次优的投资组合。 为解决这一局限性，稳健优化通过引入不确定性集来提高市场波动下的投资组合可靠性。 本研究使用移动窗口和引导方法构建这些不确定性集，并将其应用于具有不同风险规避参数的印尼银行股票数据。 结果表明，与引导方法相比，使用移动窗口方法的稳健优化，尤其是在较小的风险规避参数下，能提供更好的风险收益权衡。 这些发现突显了移动窗口方法在为风险容忍型投资者生成更有效投资策略方面的潜力。 显示英文摘要

Since the COVID-19 pandemic, the number of investors in the Indonesia Stock Exchange has steadily increased, emphasizing the importance of portfolio optimization in balancing risk and return. The classical mean-variance optimization model, while widely applied, depends on historical return and risk estimates that are uncertain and may result in suboptimal portfolios. To address this limitation, robust optimization incorporates uncertainty sets to improve portfolio reliability under market fluctuations. This study constructs such sets using moving-window and bootstrapping methods and applies them to Indonesian banking stock data with varying risk-aversion parameters. The results show that robust optimization with the moving-window method, particularly with a smaller risk-aversion parameter, provides a better risk-return trade-off compared to the bootstrapping approach. These findings highlight the potential of the moving-window method to generate more effective portfolio strategies for risk-tolerant investors.

[6] arXiv:2510.15307 [中文pdf, pdf, 其他]

标题： 学术不端行为中的战略互动：考试试卷交换机制的博弈论分析 显示英文标题

标题： Strategic Interactions in Academic Dishonesty: A Game-Theoretic Analysis of the Exam Script Swapping Mechanism

Venkat Ram Reddy Ganuthula, Manish Kumar Singh

主题： 一般经济学 (econ.GN) ; 计算机科学与博弈论 (cs.GT)

本文提出了一种新颖的博弈论框架，通过一种独特的威慑机制——被抓到抄袭的学生之间的强制考试试卷交换，来分析学术不诚实行为。 我们将学生之间的战略互动建模为一个具有不对称信息的非合作博弈，并研究了三种基础情景：不对称准备水平、相互不准备和协调的部分准备。 我们的分析表明，通过在结果中引入战略相互依赖性，试卷交换惩罚比传统惩罚产生了更强的威慑效果。 纳什均衡分析表明，相互准备成为占优策略。 该框架为制度政策设计提供了见解，表明能够创造相互脆弱性的非常规惩罚机制可能比传统的个体惩罚更有效。 提出了未来的经验验证和行为实验来测试模型预测，包括对惩罚严重性随时间减弱效应的探索。 显示英文摘要

This paper presents a novel game theoretic framework for analyzing academic dishonesty through the lens of a unique deterrent mechanism: forced exam script swapping between students caught copying. We model the strategic interactions between students as a non cooperative game with asymmetric information and examine three base scenarios asymmetric preparation levels, mutual non preparation, and coordinated partial preparation. Our analysis reveals that the script swapping punishment creates a stronger deterrent effect than traditional penalties by introducing strategic interdependence in outcomes. The Nash equilibrium analysis demonstrates that mutual preparation emerges as the dominant strategy. The framework provides insights for institutional policy design, suggesting that unconventional punishment mechanisms that create mutual vulnerability can be more effective than traditional individual penalties. Future empirical validation and behavioral experiments are proposed to test the model predictions, including explorations of tapering off effects in punishment severity over time.

[7] arXiv:2510.15399 [中文pdf, pdf, 其他]

标题： 国际迁移与留守家庭的饮食多样性：来自印度的证据 显示英文标题

标题： International migration and dietary diversity of left-behind households: evidence from India

Pooja Batra, Ajay Sharma

评论： 发表于《食品安全》

主题： 一般经济学 (econ.GN)

在本文中，我们分析了国际移民对留守家庭食物消费和饮食多样性的影响。 使用2011年喀拉拉邦移民调查数据，我们研究了因国际移民而有移民的家庭是否比没有移民的家庭具有更高的消费支出和更丰富的饮食多样性。 我们使用普通最小二乘法和工具变量方法来回答这个问题。 主要发现是：a) 移民家庭的整体消费支出以及食品支出都更高；b) 我们发现国际移民导致了留守家庭饮食多样性的增加。 此外，我们探讨了对农村和城市家庭的食品子类别的支出影响。 我们发现，移民家庭在蛋白质（牛奶、豆类和鸡蛋、鱼和肉类）上的支出更多，同时他们在不健康饮食习惯（加工食品和即食食品）上的支出也更高。 显示英文摘要

In this paper, we analyse the impact of international migration on the food consumption and dietary diversity of left-behind households. Using the Kerala migration survey 2011, we study whether households with emigrants (on account of international migration) have higher consumption expenditure and improved dietary diversity than their non-migrating counterparts. We use ordinary least square and instrumental variable approach to answer this question. The key findings are that: a) emigrant households have higher overall consumption expenditure as well as higher expenditure on food; b) we find that international migration leads to increase in the dietary diversity of left behind households. Further, we explore the effect on food sub-group expenditure for both rural and urban households. We find that emigrant households spend more on protein (milk, pulses and egg, fish and meat), at the same time there is higher spending on non-healthy food habits (processed and ready to eat food items) among them.

[8] arXiv:2510.15405 [中文pdf, pdf, 其他]

标题： 三点规则变化对足球比赛中竞争平衡的影响：合成控制方法的方法论 显示英文标题

标题： Impact of Three-Point Rule Change on Competitive Balance in Football: A Synthetic Control Method Approach

Ajay Sharma

评论： 发表于《应用经济学》

主题： 一般经济学 (econ.GN)

体育管理机构经常对规则和规定进行更改以提高竞争性。 1981年，英格兰足球协会做出了一项这样的更改，当时它将国内联赛中胜场积分的规则从2分制改为3分制。 本研究旨在通过合成控制方法的准实验估计设计，衡量这一规则变化对国内联赛竞争平衡的影响。 3分制的改变导致了英格兰联赛竞争平衡的增加。 此外，我们显示每场比赛进球数没有显著变化。 显示英文摘要

Governing authorities in sports often make changes to rules and regulations to increase competitiveness. One such change was made by the English Football Association in 1981 when it changed the rule for awarding points in the domestic league from two points for a win to three points. This study aims to measure this rule change's impact on the domestic league's competitive balance using a quasi-experimental estimation design of a synthetic control method. The three-point rule change led to an increase in competitive balance in the English League. Further, we show no significant change in the number of goals scored per match.

[9] arXiv:2510.15420 [中文pdf, pdf, 其他]

标题： 移民之间的异质性、教育与职业不匹配以及教育回报：来自印度的证据 显示英文标题

标题： Heterogeneity among migrants, education-occupation mis-match and returns to education: Evidence from India

Shweta Bahl, Ajay Sharma

评论： 发表于区域研究

主题： 一般经济学 (econ.GN)

使用印度全国有代表性的数据，本文研究了内部迁移者在教育职业不匹配情况以及教育回报和EOM的情况，并考虑了他们之间的异质性。 特别是，本研究考虑了由于迁移原因、人口特征、空间因素、迁移经验和迁移类型而产生的异质性。 分析表明，根据迁移原因、人口特征和空间因素的不同，EOM的发生率和回报率存在差异。 该研究强调了关注EOM的必要性，以提高迁移的生产率效益。 它还提供了减少迁移者被错配的可能性同时最大化其教育回报的框架。 显示英文摘要

Using nationally representative data for India, this paper examines the incidence of education occupation mismatch and returns to education and EOM for internal migrants while considering the heterogeneity among them. In particular, this study considers heterogeneity arising because of the reason to migrate, demographic characteristics, spatial factors, migration experience, and type of migration. The analysis reveals that there exists variation in the incidence and returns to EOM depending on the reason to migrate, demographic characteristics, and spatial factors. The study highlights the need of focusing on EOM to increase the productivity benefits of migration. It also provides the framework for minimizing migrants' likelihood of being mismatched while maximizing their returns to education.

[10] arXiv:2510.15617 [中文pdf, pdf, html, 其他]

标题： 政治干预减少一次性塑料（SUPs）和价格效应：奥地利和德国的事件研究 显示英文标题

标题： Political Interventions to Reduce Single-Use Plastics (SUPs) and Price Effects: An Event Study for Austria and Germany

Felix Reichel

评论： 11页，4图，2表，13参考文献，1附录

主题： 一般经济学 (econ.GN)

一次性塑料（SUPs）造成巨大的环境成本。 在《指令》（EU）2019/904之后，奥地利和德国引入了生产者收费和基金支付，旨在覆盖清理工作。 使用包含价格的高频零售报价面板数据，并采用双向聚类标准误差的固定效应事件研究，本文衡量这些成本对消费者价格的影响程度。 我们发现奥地利有明显的价格传递。 当将奥地利产品合并时，处理项目在十二个月内比非SUP对照组高13.01个指数点（DiD(12m)；p<0.001），在整个后续期间高19.42个点（p<0.001）。 按产品来看，气球显示出强烈且持久的效果（DiD(12m)=13.43，p=0.007；Full DiD=19.96，p<0.001）。 杯子在短期内表现出混合的变动（例如，DiD(12m)=-22.73，p=0.096），并在整个期间表现出积极但不精确的对比。 显示英文摘要

Single-use plastics (SUPs) create large environmental costs. After Directive (EU) 2019/904, Austria and Germany introduced producer charges and fund payments meant to cover clean-up work. Using a high-frequency panel of retail offer spells containing prices and a fixed-effects event study with two-way clustered standard errors, this paper measures how much these costs drive up consumer prices. We find clear price pass-through in Austria. When Austrian products are pooled, treated items are 13.01 index points higher than non-SUP controls within twelve months (DiD(12m); p<0.001) and 19.42 points over the full post period (p<0.001). By product, balloons show strong and lasting effects (DiD(12m)=13.43, p=0.007; Full DiD=19.96, p<0.001). Cups show mixed short-run movements (e.g., DiD(12m)=-22.73, p=0.096) and a positive but imprecise full-period contrast.

[11] arXiv:2510.15691 [中文pdf, pdf, 其他]

标题： 探索定量因素与大型语言模型中的新闻流表示之间的协同作用用于股票收益预测 显示英文标题

标题： Exploring the Synergy of Quantitative Factors and Newsflow Representations from Large Language Models for Stock Return Prediction

Tian Guo, Emmanuel Hauptmann

主题： 计算金融 (q-fin.CP) ; 人工智能 (cs.AI) ; 计算与语言 (cs.CL) ; 机器学习 (cs.LG)

在量化投资中，收益预测支持各种任务，包括股票选择、投资组合优化和风险管理。 量化因子，如估值、质量和增长，捕捉股票的各种特征。 非结构化金融数据，如新闻和电话会议记录，由于大型语言模型（LLMs）的最新进展而受到越来越多的关注。 本文研究了在收益预测和股票选择中利用多模态因子和新闻流的有效方法。 首先，我们引入一种融合学习框架，从由LLM生成的因子和新闻流表示中学习统一表示。 在此框架内，我们比较了三种代表性方法：表示组合、表示求和和注意表示。 接下来，基于融合学习中的实证观察，我们探索了一种混合模型，该模型自适应地结合单模态及其融合的预测结果。 为了缓解混合模型中观察到的训练不稳定问题，我们引入了一种具有理论见解的解耦训练方法。 最后，我们在实际投资范围内进行的实验得出了关于因子和新闻在股票收益预测中有效多模态建模的几个见解。 显示英文摘要

In quantitative investing, return prediction supports various tasks, including stock selection, portfolio optimization, and risk management. Quantitative factors, such as valuation, quality, and growth, capture various characteristics of stocks. Unstructured financial data, like news and transcripts, has attracted growing attention, driven by recent advances in large language models (LLMs). This paper examines effective methods for leveraging multimodal factors and newsflow in return prediction and stock selection. First, we introduce a fusion learning framework to learn a unified representation from factors and newsflow representations generated by an LLM. Within this framework, we compare three representative methods: representation combination, representation summation, and attentive representations. Next, building on empirical observations from fusion learning, we explore the mixture model that adaptively combines predictions made by single modalities and their fusion. To mitigate the training instability observed in the mixture model, we introduce a decoupled training approach with theoretical insights. Finally, our experiments on real investment universes yield several insights into effective multimodal modeling of factors and news for stock return prediction.

[12] arXiv:2510.15709 [中文pdf, pdf, html, 其他]

标题： 稳健的保险定价与流动性管理 显示英文标题

标题： Robust Insurance Pricing and Liquidity Management

Shunzhi Pang

主题： 风险管理 (q-fin.RM)

随着新兴风险的增加，模型不确定性在保险行业中构成了一个根本性的挑战，使得稳健定价成为一个首要问题。 本文研究了保险公司对稳健性的偏好如何塑造动态保险市场中的竞争均衡。 保险公司优化其承保和流动性管理策略以最大化股东价值，从而导致可以解析推导和数值求解的均衡结果。 与没有模型不确定性的基准相比，稳健的保险定价导致了显著更高的保费和股权估值。 值得注意的是，我们的模型得出了三个新的见解：（1）总承保能力的最小值、最大值和可接受范围都扩大了，表明保险公司的流动性管理变得更加保守。 （2）承保周期的预期长度显著增加，远远超过早期实证研究中通常报告的范围。 （3）虽然在长期中容量过程仍然是遍历的，但平稳密度在低容量状态中更加集中，这意味着流动性受限的保险公司需要更长时间才能恢复。 这些发现共同为近期对承保周期实证证据的怀疑提供了一个可能的解释，表明这些周期确实存在，但比之前假设的要长得多。 显示英文摘要

With the rise of emerging risks, model uncertainty poses a fundamental challenge in the insurance industry, making robust pricing a first-order question. This paper investigates how insurers' robustness preferences shape competitive equilibrium in a dynamic insurance market. Insurers optimize their underwriting and liquidity management strategies to maximize shareholder value, leading to equilibrium outcomes that can be analytically derived and numerically solved. Compared to a benchmark without model uncertainty, robust insurance pricing results in significantly higher premiums and equity valuations. Notably, our model yields three novel insights: (1) The minimum, maximum, and admissible range of aggregate capacity all expand, indicating that insurers' liquidity management becomes more conservative. (2) The expected length of the underwriting cycle increases substantially, far exceeding the range commonly reported in earlier empirical studies. (3) While the capacity process remains ergodic in the long run, the stationary density becomes more concentrated in low-capacity states, implying that liquidity-constrained insurers require longer to recover. Together, these findings provide a potential explanation for recent skepticism regarding the empirical evidence of underwriting cycles, suggesting that such cycles may indeed exist but are considerably longer than previously assumed.

交叉提交 (展示 6 之 6 条目 )

[13] arXiv:2510.15205 (交叉列表自 cs.CE) [中文pdf, pdf, html, 其他]

标题： 向预测市场的黑-舒尔斯模型：统一核与市场制造者手册 显示英文标题

标题： Toward Black Scholes for Prediction Markets: A Unified Kernel and Market Maker's Handbook

Shaw Dalen

主题： 计算工程、金融与科学 (cs.CE) ; 计算金融 (q-fin.CP)

预测市场，如Polymarket，将分散的信息聚合为可交易的概率，但它们仍然缺乏类似于期权从Black-Scholes获得的统一随机核。 随着这些市场随着机构参与、交易所集成以及选举和宏观数据发布期间的更高交易量而扩展，做市商面临信念波动、跳跃和跨事件风险，而没有标准化的报价或对冲工具。 我们提出这样的基础：一种具有风险中性漂移的逻辑跳跃扩散模型，该模型将交易概率p_t视为Q鞅，并将信念波动、跳跃强度和依赖性作为可报价的风险因素。 在此基础上，我们构建了一个校准流程，该流程过滤微观结构噪声，使用期望最大化分离扩散和跳跃，强制风险中性漂移，并产生一个稳定的信念波动曲面。 然后，我们定义了一个连贯的衍生层（方差、相关性、走廊和首次通过工具），类似于期权市场中的波动率和相关性产品。 在合成风险中性路径和真实事件数据的受控实验中，该模型相对于仅扩散和概率空间基线，减少了短期信念方差预测误差，支持因果校准和经济可解释性。 从概念上讲，逻辑跳跃扩散内核为预测市场提供了隐含波动率的类比：一种可用于报价、对冲和在诸如Polymarket等场所转移信念风险的可处理且可交易的语言。 显示英文摘要

Prediction markets, such as Polymarket, aggregate dispersed information into tradable probabilities, but they still lack a unifying stochastic kernel comparable to the one options gained from Black-Scholes. As these markets scale with institutional participation, exchange integrations, and higher volumes around elections and macro prints, market makers face belief volatility, jump, and cross-event risks without standardized tools for quoting or hedging. We propose such a foundation: a logit jump-diffusion with risk-neutral drift that treats the traded probability p_t as a Q-martingale and exposes belief volatility, jump intensity, and dependence as quotable risk factors. On top, we build a calibration pipeline that filters microstructure noise, separates diffusion from jumps using expectation-maximization, enforces the risk-neutral drift, and yields a stable belief-volatility surface. We then define a coherent derivative layer (variance, correlation, corridor, and first-passage instruments) analogous to volatility and correlation products in option markets. In controlled experiments on synthetic risk-neutral paths and real event data, the model reduces short-horizon belief-variance forecast error relative to diffusion-only and probability-space baselines, supporting both causal calibration and economic interpretability. Conceptually, the logit jump-diffusion kernel supplies an implied-volatility analogue for prediction markets: a tractable, tradable language for quoting, hedging, and transferring belief risk across venues such as Polymarket.

[14] arXiv:2510.15423 (交叉列表自 math.PR) [中文pdf, pdf, html, 其他]

标题： 利用马里亚文微积分的看涨障碍期权短期行为分析 显示英文标题

标题： On the short-time behaviour of up-and-in barrier options using Malliavin calculus

Òscar Burés

评论： 21页，3图

主题： 概率 (math.PR) ; 数学金融 (q-fin.MF)

在本文中，我们研究了在广泛一类随机波动率模型下，向上敲入障碍期权的短期到期渐近行为。我们的方法使用了通常用于线性随机偏微分方程的马尔可夫微分技巧，来分析对数价格过程的最大值的分布。我们推导了一个集中不等式，并以到期时间表示最大值密度的显式边界。这些结果给出了当到期时间消失时，向上敲入障碍期权价格的渐近衰减率的上界。我们进一步展示了我们框架在粗糙伯格米模型中的适用性，并通过数值实验验证了理论结果。 显示英文摘要

In this paper we study the short-maturity asymptotics of up-and-in barrier options under a broad class of stochastic volatility models. Our approach uses Malliavin calculus techniques, typically used for linear stochastic partial differential equations, to analyse the law of the supremum of the log-price process. We derive a concentration inequality and explicit bounds on the density of the supremum in terms of the time to maturity. These results yield an upper bound on the asymptotic decay rate of up-and-in barrier option prices as maturity vanishes. We further demonstrate the applicability of our framework to the rough Bergomi model and validate the theoretical results with numerical experiments.

[15] arXiv:2510.15458 (交叉列表自 stat.ML) [中文pdf, pdf, html, 其他]

标题： 因果模型中的鲁棒优化和G-因果归一化流 显示英文标题

标题： Robust Optimization in Causal Models and G-Causal Normalizing Flows

Gabriele Visentin, Patrick Cheridito

主题： 机器学习 (stat.ML) ; 人工智能 (cs.AI) ; 机器学习 (cs.LG) ; 投资组合管理 (q-fin.PM)

在本文中，我们表明，在因果模型中的干预鲁棒优化问题在$G$-因果 Wasserstein 距离下是连续的，但在标准 Wasserstein 距离下可能不连续。这突出了在为这类任务增强数据时使用尊重因果结构的生成模型的重要性。为此，我们提出了一种新的归一化流架构，该架构满足因果结构模型的通用逼近性质，并可以高效地训练以最小化$G$-因果 Wasserstein 距离。实证上，我们展示了我们的模型在因果回归和因果因子模型中的均值-方差投资组合优化的数据增强任务中优于标准（非因果）生成模型。 显示英文摘要

In this paper, we show that interventionally robust optimization problems in causal models are continuous under the $G$-causal Wasserstein distance, but may be discontinuous under the standard Wasserstein distance. This highlights the importance of using generative models that respect the causal structure when augmenting data for such tasks. To this end, we propose a new normalizing flow architecture that satisfies a universal approximation property for causal structural models and can be efficiently trained to minimize the $G$-causal Wasserstein distance. Empirically, we demonstrate that our model outperforms standard (non-causal) generative models in data augmentation for causal regression and mean-variance portfolio optimization in causal factor models.

[16] arXiv:2510.15509 (交叉列表自 cs.CY) [中文pdf, pdf, html, 其他]

标题： 人工智能在非政府组织中的采用：系统文献综述 显示英文标题

标题： AI Adoption in NGOs: A Systematic Literature Review

Janne Rotter, William Bailkoski

主题： 计算机与社会 (cs.CY) ; 人工智能 (cs.AI) ; 一般经济学 (econ.GN)

人工智能有潜力显著改善非政府组织如何利用有限的资源实现社会利益，但关于非政府组织如何采用人工智能的证据仍然零散。 在本研究中，我们系统地调查了非政府组织中人工智能采用的使用案例类型，并识别出常见的挑战和解决方案，这些挑战和解决方案基于组织规模和地理背景进行情境化分析。 我们回顾了现有的原始文献，包括2020年至2025年间以英文发表的与非政府组织人工智能采用相关的社会影响研究。 按照PRISMA协议，两名独立审稿人进行研究选择，并定期交叉检查以确保方法学严谨性，最终形成了包含65项研究的文献库。 利用主题和叙述的方法，我们识别出非政府组织中的六种人工智能使用案例类别——参与、创造力、决策、预测、管理与优化，并在技术-组织-环境（TOE）框架内提取了常见的挑战和解决方案。 通过整合我们的发现，这项综述为非政府组织的人工智能采用提供了新的理解，将特定的使用案例和挑战与组织和环境因素联系起来。 我们的结果表明，尽管人工智能前景广阔，但非政府组织的人工智能采用仍然不均衡，并偏向于大型组织。 然而，遵循基于文献的路线图可以帮助非政府组织克服人工智能采用的初始障碍，最终提高效率、参与度和社会影响。 显示英文摘要

AI has the potential to significantly improve how NGOs utilize their limited resources for societal benefits, but evidence about how NGOs adopt AI remains scattered. In this study, we systematically investigate the types of AI adoption use cases in NGOs and identify common challenges and solutions, contextualized by organizational size and geographic context. We review the existing primary literature, including studies that investigate AI adoption in NGOs related to social impact between 2020 and 2025 in English. Following the PRISMA protocol, two independent reviewers conduct study selection, with regular cross-checking to ensure methodological rigour, resulting in a final literature body of 65 studies. Leveraging a thematic and narrative approach, we identify six AI use case categories in NGOs - Engagement, Creativity, Decision-Making, Prediction, Management, and Optimization - and extract common challenges and solutions within the Technology-Organization-Environment (TOE) framework. By integrating our findings, this review provides a novel understanding of AI adoption in NGOs, linking specific use cases and challenges to organizational and environmental factors. Our results demonstrate that while AI is promising, adoption among NGOs remains uneven and biased towards larger organizations. Nevertheless, following a roadmap grounded in literature can help NGOs overcome initial barriers to AI adoption, ultimately improving effectiveness, engagement, and social impact.

[17] arXiv:2510.15612 (交叉列表自 cs.CE) [中文pdf, pdf, html, 其他]

标题： SoK：去中心化预测市场（DePMs）的市场微观结构 显示英文标题

标题： SoK: Market Microstructure for Decentralized Prediction Markets (DePMs)

Nahid Rahman, Joseph Al-Chami, Jeremy Clark

主题： 计算工程、金融与科学 (cs.CE) ; 密码学与安全 (cs.CR) ; 交易与市场微观结构 (q-fin.TR)

去中心化预测市场（DePMs）允许在基于事件的投注中开放参与，而无需完全依赖中心化中介。 我们回顾了DePMs的历史，其可追溯至2011年，并包括数百个提案。 也许令人惊讶的是，现代DePM如Polymarket与早期设计如Truthcoin和Augur v1存在显著差异。 我们通过回顾提出一个由七个阶段组成的模块化工作流程：底层基础设施、市场主题、股份结构和定价、交易、市场解决、结算和归档。 对于每个模块，我们列举了设计变体，并分析了在去中心化、表达能力和抗操纵性方面的权衡。 我们还识别了对该生态系统感兴趣的研究人员的开放问题。 显示英文摘要

Decentralized prediction markets (DePMs) allow open participation in event-based wagering without fully relying on centralized intermediaries. We review the history of DePMs which date back to 2011 and includes hundreds of proposals. Perhaps surprising, modern DePMs like Polymarket deviate materially from earlier designs like Truthcoin and Augur v1. We use our review to present a modular workflow comprising seven stages: underlying infrastructure, market topic, share structure and pricing, trading, market resolution, settlement, and archiving. For each module, we enumerate the design variants, analyzing trade-offs around decentralization, expressiveness, and manipulation resistance. We also identify open problems for researchers interested in this ecosystem.

[18] arXiv:2510.15616 (交叉列表自 math.PR) [中文pdf, pdf, html, 其他]

标题： 鞅理论在具有不对称信息的迪尼游戏中的应用 显示英文标题

标题： Martingale theory for Dynkin games with asymmetric information

Tiziano De Angelis, Jan Palczewski, Jacob Smith

评论： 69页

主题： 概率 (math.PR) ; 优化与控制 (math.OC) ; 数学金融 (q-fin.MF)

本文提供了随机停止时间对形成零和Dynkin博弈鞍点的必要且充分条件，该博弈在参与者之间具有部分和/或不对称信息。 框架是非马尔可夫的，并涵盖了基本上任何信息结构。 我们的方法依赖于涉及参与者均衡收益的适当上鞅和下鞅的识别。 鞍点策略根据这些均衡收益的动力学进行表征，并与它们的Doob-Meyer分解相关。 显示英文摘要

This paper provides necessary and sufficient conditions for a pair of randomised stopping times to form a saddle point of a zero-sum Dynkin game with partial and/or asymmetric information across players. The framework is non-Markovian and covers essentially any information structure. Our methodology relies on the identification of suitable super and submartingales involving players' equilibrium payoffs. Saddle point strategies are characterised in terms of the dynamics of those equilibrium payoffs and are related to their Doob-Meyer decompositions.

替换提交 (展示 7 之 7 条目 )

[19] arXiv:2402.17523 (替换) [中文pdf, pdf, html, 其他]

标题： 高维投资组合分析与TE和权重约束 显示英文标题

标题： Portfolio Analysis in High Dimensions with TE and Weight Constraints

Mehmet Caner, Qingliang Fan

主题： 投资组合管理 (q-fin.PM) ; 统计金融 (q-fin.ST)

本文探讨了在高维资产集合中形成约束最优投资组合的统计特性。 我们研究了具有跟踪误差约束的投资组合，同时具有跟踪误差和权重限制的投资组合，以及仅受权重限制的投资组合。 跟踪误差衡量投资组合相对于基准（通常是指数）的表现，而权重约束则根据监管要求或基金说明书确定资产配置。 我们的方法采用了一种新颖的统计学习技术，将因子模型与节点回归相结合，称为受限残差节点最优权重回归（CROWN）方法。 我们展示了其在大维度下的估计一致性，即使资产数量超过投资组合的时间跨度。 提供了约束投资组合权重、风险和夏普比率的收敛速率结果，模拟和实证证据突显了该方法的卓越性能。 显示英文摘要

This paper explores the statistical properties of forming constrained optimal portfolios within a high-dimensional set of assets. We examine portfolios with tracking error constraints, those with simultaneous tracking error and weight restrictions, and portfolios constrained solely by weight. Tracking error measures portfolio performance against a benchmark (typically an index), while weight constraints determine asset allocation based on regulatory requirements or fund prospectuses. Our approach employs a novel statistical learning technique that integrates factor models with nodewise regression, named the Constrained Residual Nodewise Optimal Weight Regression (CROWN) method. We demonstrate its estimation consistency in large dimensions, even when assets outnumber the portfolio's time span. Convergence rate results for constrained portfolio weights, risk, and Sharpe Ratio are provided, and simulation and empirical evidence highlight the method's outstanding performance.

[20] arXiv:2408.17177 (替换) [中文pdf, pdf, html, 其他]

标题： 狼人杀游戏中的最优策略：理论研究 显示英文标题

标题： Optimal Strategy in the Werewolf Game: A Theoretical Study

ST Wang

主题： 一般经济学 (econ.GN)

在本文中，我们从博弈论的角度研究了狼人杀游戏——一种广泛进行的战略社会推理游戏，涉及两个对立派系的最优策略。 我们考虑两种情景：没有先知的游戏中和有先知的游戏中。 在没有先知的情景中，我们提出了一种增强策略，称为“随机策略+”，该策略显著提高了狼人组相对于传统随机策略的获胜概率。 在有先知的情景中，我们将游戏重新建模为一个特定约束下的扩展形式贝叶斯博弈，并推导出先知的最优策略，该策略导致了一个完美贝叶斯均衡（PBE）。 本研究为建模狼人杀游戏提供了一个严格的分析框架，并对在不对称和不完全信息下战略决策制定提供了更广泛的见解。 显示英文摘要

In this paper, we investigate the optimal strategies in the Werewolf Game-a widely played strategic social deduction game involving two opposing factions-from a game-theoretic perspective. We consider two scenarios: the game without a prophet and the game with a prophet. In the scenario without a prophet, we propose an enhanced strategy called ``random strategy+'' that significantly improves the werewolf group's winning probability over conventional random strategies. In the scenario with a prophet, we reformulate the game as an extensive-form Bayesian game under a specific constraint, and derive the prophet's optimal strategy that induces a Perfect Bayesian Equilibrium (PBE). This study provides a rigorous analytical framework for modeling the Werewolf Game and offers broader insights into strategic decision-making under asymmetric and incomplete information.

[21] arXiv:2409.17035 (替换) [中文pdf, pdf, html, 其他]

标题： 扩展到云：云技术在小公司和大公司中的使用和增长率 显示英文标题

标题： Scaling up to the cloud: Cloud technology use and growth rates in small and large firms

Bernardo Caldarola, Luca Fontanelli

主题： 一般经济学 (econ.GN)

最近的实证证据表明，对信息和通信技术（ICT）的投资在改善大公司绩效方面比小公司更为显著。 然而，ICT可能并不相同，因为它们对公司在组织结构上的影响各不相同。 我们研究了云服务的使用对法国公司长期规模增长率的影响。 我们发现云服务对公司的增长速度有积极影响，小公司相比大公司更能享受到显著的好处。 我们的研究结果表明，云技术有助于降低数字化的障碍，这些障碍尤其影响小公司。 通过降低这些障碍，云技术的采用提高了可扩展性并释放了未开发的增长潜力。 显示英文摘要

Recent empirical evidence shows that investments in ICT disproportionately improve the performance of larger firms versus smaller ones. However, ICT may not be all alike, as they differ in their impact on firms' organisational structure. We investigate the effect of the use of cloud services on the long run size growth rate of French firms. We find that cloud services positively impact firms' growth rates, with smaller firms experiencing more significant benefits compared to larger firms. Our findings suggest cloud technologies help reduce barriers to digitalisation, which affect especially smaller firms. By lowering these barriers, cloud adoption enhances scalability and unlocks untapped growth potential.

[22] arXiv:2506.18147 (替换) [中文pdf, pdf, html, 其他]

标题： 因果干预在债券多交易商对客户平台中 显示英文标题

标题： Causal Interventions in Bond Multi-Dealer-to-Client Platforms

Paloma Marín, Sergio Ardanza-Trevijano, Javier Sabio

主题： 交易与市场微观结构 (q-fin.TR)

金融市场的数字化已将交易从语音渠道转移到电子渠道，现在多做市商对客户（MD2C）平台使客户能够同时向多个做市商请求金融工具（如债券）的报价（RfQ）。 在这一竞争环境中，做市商无法看到彼此的价格，因此对谈判过程进行严格的分析对于确保其盈利能力至关重要。 本文介绍了一种使用概率图模型和因果推断分析RfQ过程的新通用框架。 在这个框架中，我们探讨了与参与MD2C平台的做市商相关的不同推断问题，例如计算最优价格、估计潜在收入以及识别可能对做市商的轴心感兴趣的客户。 然后我们转向分析两种不同的模型规范方法：一种是基于（Fermanian, Guéant, & Pu, 2017）工作的生成模型；另一种是利用机器学习技术的判别模型。 我们的结果表明，生成模型可以在保持预测准确性的同时满足关键的业务要求，例如价差单调性，其预测准确性可以与领先的判别算法如LightGBM（ROC-AUC：0.742 vs. 0.743）相媲美。 显示英文摘要

The digitalization of financial markets has shifted trading from voice to electronic channels, with Multi-Dealer-to-Client (MD2C) platforms now enabling clients to request quotes (RfQs) for financial instruments like bonds from multiple dealers simultaneously. In this competitive landscape, dealers cannot see each other's prices, making a rigorous analysis of the negotiation process crucial to ensure their profitability. This article introduces a novel general framework for analyzing the RfQ process using probabilistic graphical models and causal inference. Within this framework, we explore different inferential questions that are relevant for dealers participating in MD2C platforms, such as the computation of optimal prices, estimating potential revenues and the identification of clients that might be interested in trading the dealer's axes. We then move into analyzing two different approaches for model specification: a generative model built on the work of (Fermanian, Gu\'eant, \& Pu, 2017); and discriminative models utilizing machine learning techniques. Our results show that generative models can match the predictive accuracy of leading discriminative algorithms such as LightGBM (ROC-AUC: 0.742 vs. 0.743) while simultaneously enforcing critical business requirements, notably spread monotonicity.

[23] arXiv:2507.20957 (替换) [中文pdf, pdf, html, 其他]

标题： 你的AI，不是你的观点：LLMs在投资分析中的偏差 显示英文标题

标题： Your AI, Not Your View: The Bias of LLMs in Investment Analysis

Hoyoung Lee, Junhyuk Seo, Suhwan Park, Junhyeong Lee, Wonbin Ahn, Chanyeol Choi, Alejandro Lopez-Lira, Yongjae Lee

评论： 被ACM国际金融人工智能会议（ICAIF）接受

主题： 投资组合管理 (q-fin.PM) ; 人工智能 (cs.AI) ; 计算与语言 (cs.CL)

在金融领域，大型语言模型（LLMs）经常面临由于预训练参数知识与实时市场数据之间的差异而产生的知识冲突。 这些冲突在现实世界的投资服务中尤其成问题，因为模型的内在偏见可能与机构目标不一致，导致不可靠的建议。 尽管存在这种风险，LLM 的内在投资偏见仍鲜有研究。 我们提出一个实验框架，以研究此类冲突场景中的涌现行为，并提供基于 LLM 的投资分析中的偏见定量分析。 通过使用平衡和不平衡论点的假设情景，我们提取模型的潜在偏见并衡量其持续性。 我们的分析集中在行业、规模和动量上，揭示了不同的、特定于模型的偏见。 在大多数模型中，观察到偏好科技股、大盘股和逆向策略的趋势。 这些基础偏见往往会升级为确认偏误，导致模型在面对越来越多的反面证据时仍然坚持最初的判断。 在 https://linqalpha.com/leaderboard 上有一个公开的排行榜，用于对更广泛的模型集进行偏差基准测试。 显示英文摘要

In finance, Large Language Models (LLMs) face frequent knowledge conflicts arising from discrepancies between their pre-trained parametric knowledge and real-time market data. These conflicts are especially problematic in real-world investment services, where a model's inherent biases can misalign with institutional objectives, leading to unreliable recommendations. Despite this risk, the intrinsic investment biases of LLMs remain underexplored. We propose an experimental framework to investigate emergent behaviors in such conflict scenarios, offering a quantitative analysis of bias in LLM-based investment analysis. Using hypothetical scenarios with balanced and imbalanced arguments, we extract the latent biases of models and measure their persistence. Our analysis, centered on sector, size, and momentum, reveals distinct, model-specific biases. Across most models, a tendency to prefer technology stocks, large-cap stocks, and contrarian strategies is observed. These foundational biases often escalate into confirmation bias, causing models to cling to initial judgments even when faced with increasing counter-evidence. A public leaderboard benchmarking bias across a broader set of models is available at https://linqalpha.com/leaderboard

[24] arXiv:2506.09080 (替换) [中文pdf, pdf, html, 其他]

标题： FinHEAR：金融决策中的人类专业知识和自适应风险感知时间推理 显示英文标题

标题： FinHEAR: Human Expertise and Adaptive Risk-Aware Temporal Reasoning for Financial Decision-Making

Jiaxiang Chen, Mingxi Zou, Zhuo Wang, Qifan Wang, Dongning Sun, Chi Zhang, Zenglin Xu

主题： 机器学习 (cs.LG) ; 人工智能 (cs.AI) ; 计算金融 (q-fin.CP)

金融决策制定对语言模型提出了独特的挑战，需要时间推理、适应性风险评估以及对动态事件的响应能力。 虽然大型语言模型（LLMs）表现出强大的通用推理能力，但它们往往无法捕捉到人类金融决策中的行为模式——例如在信息不对称情况下的专家依赖、损失厌恶敏感性和反馈驱动的时间调整。 我们提出了FinHEAR，一个用于人类专业知识和自适应风险感知推理的多智能体框架。 FinHEAR协调专门的基于LLM的代理，在以事件为中心的流程中分析历史趋势、解释当前事件并检索专家指导的先例。 基于行为经济学，它结合了专家引导的检索、置信度调整的头寸大小和基于结果的优化，以提高可解释性和鲁棒性。 在整理好的金融数据集上的实证结果表明，FinHEAR在趋势预测和交易任务中始终优于强基线，实现了更高的准确率和更好的风险调整回报。 显示英文摘要

Financial decision-making presents unique challenges for language models, demanding temporal reasoning, adaptive risk assessment, and responsiveness to dynamic events. While large language models (LLMs) show strong general reasoning capabilities, they often fail to capture behavioral patterns central to human financial decisions-such as expert reliance under information asymmetry, loss-averse sensitivity, and feedback-driven temporal adjustment. We propose FinHEAR, a multi-agent framework for Human Expertise and Adaptive Risk-aware reasoning. FinHEAR orchestrates specialized LLM-based agents to analyze historical trends, interpret current events, and retrieve expert-informed precedents within an event-centric pipeline. Grounded in behavioral economics, it incorporates expert-guided retrieval, confidence-adjusted position sizing, and outcome-based refinement to enhance interpretability and robustness. Empirical results on curated financial datasets show that FinHEAR consistently outperforms strong baselines across trend prediction and trading tasks, achieving higher accuracy and better risk-adjusted returns.

[25] arXiv:2506.15305 (替换) [中文pdf, pdf, html, 其他]

标题： 供应链金融中的条件生成建模以增强信用风险管理 显示英文标题

标题： Conditional Generative Modeling for Enhanced Credit Risk Management in Supply Chain Finance

Qingkai Zhang, L. Jeff Hong, Houmin Yan

主题： 机器学习 (cs.LG) ; 风险管理 (q-fin.RM)

跨境电子商务（CBEC）的迅速扩展为中小型卖家创造了重大机遇，但由于他们的信用历史有限，融资仍然是一个关键挑战。 第三方物流（3PL）主导的供应链金融（SCF）已成为一种有前景的解决方案，利用在途库存作为抵押品。 我们提出了一种针对3PL主导的SCF的先进信用风险管理框架，解决了信用风险评估和贷款规模确定的双重挑战。 具体而言，我们通过基于分位数回归的生成元模型（QRGMM）对销售分布进行条件生成建模，作为风险度量估计的基础。 我们提出了一种统一的框架，能够在引入功能风险度量公式的同时，灵活地估计多个风险度量，该公式系统地捕捉这些风险度量与不同贷款水平之间的关系，并得到理论保证的支持。 为了捕捉电子商务销售数据中的复杂协变量交互，我们将QRGMM与深度因子分解机（DeepFM）相结合。 在合成数据和真实数据上的大量实验验证了我们的模型在信用风险评估和贷款规模确定方面的有效性。 本研究探讨了生成模型在CBEC SCF风险管理中的应用，展示了它们在加强信用评估和支持中小型卖家融资方面的潜力。 显示英文摘要

The rapid expansion of cross-border e-commerce (CBEC) has created significant opportunities for small- and medium-sized sellers, yet financing remains a critical challenge due to their limited credit histories. Third-party logistics (3PL)-led supply chain finance (SCF) has emerged as a promising solution, leveraging in-transit inventory as collateral. We propose an advanced credit risk management framework tailored for 3PL-led SCF, addressing the dual challenges of credit risk assessment and loan size determination. Specifically, we leverage conditional generative modeling of sales distributions through Quantile-Regression-based Generative Metamodeling (QRGMM) as the foundation for risk measures estimation. We propose a unified framework that enables flexible estimation of multiple risk measures while introducing a functional risk measure formulation that systematically captures the relationship between these risk measures and varying loan levels, supported by theoretical guarantees. To capture complex covariate interactions in e-commerce sales data, we integrate QRGMM with Deep Factorization Machines (DeepFM). Extensive experiments on synthetic and real-world data validate the efficacy of our model for credit risk assessment and loan size determination. This study explores the use of generative models in CBEC SCF risk management, illustrating their potential to strengthen credit assessment and support financing for small- and medium-sized sellers.