Large Language Models (LLMs) are increasingly integrated into financial workflows, but evaluation practice has not kept up. Finance-specific biases can inflate performance, contaminate backtests, and make reported results useless for any deployment claim. We identify five recurring biases in financial LLM applications. They include look-ahead bias, survivorship bias, narrative bias, objective bias, and cost bias. These biases break financial tasks in distinct ways and they often compound to create an illusion of validity. We reviewed 164 papers from 2023 to 2025 and found that no single bias is discussed in more than 28 percent of studies. This position paper argues that bias in financial LLM systems requires explicit attention and that structural validity should be enforced before any result is used to support a deployment claim. We propose a Structural Validity Framework and an evaluation checklist with minimal requirements for bias diagnosis and future system design. The material is available at https://github.com/Eleanorkong/Awesome-Financial-LLM-Bias-Mitigation.
Fabio Bagarello, Francesco Gargano, Polina Khrennikova
We consider state of the art applications of artificial intelligence (AI) in modelling human financial expectations and explore the potential of quantum logic to drive future advancements in this field. This analysis highlights the application of machine learning techniques, including reinforcement learning and deep neural networks, in financial statement analysis, algorithmic trading, portfolio management, and robo-advisory services. We further discuss the emergence and progress of quantum machine learning (QML) and advocate for broader exploration of the advantages provided by quantum-inspired neural networks.
I present data from SCImago, 1999-2024, on factors that affect the valuation of finance journals. The SCImago data show some interesting facts; for example, in 2024, Finance Research Letters, valued at \$16.5 million, was four times more valuable than the Journal of Financial Economics (\$4.1m), five times more valuable than the Review of Financial Studies (\$3.3m), and over seven times more valuable than the Journal of Finance (\$2.3m). The data show that quality, as proxied by citations, positively affects journal value, and this value increase is heightened by citations from policy documents; similarly, the absence of quality, or a greater percentage of uncited documents, negatively affects journal value. The data indicate that the creation of the Elsevier Ecosystem of networked finance journals significantly enhances journal value. Further, a greater percentage of female authors positively impacts value. I discuss policy implications and suggestions for future work based on the evidence presented.
Abstract One of the most notable developments in the asset management industry in recent decades has been the growth of algorithmic trading. At the same time, significant structural changes in the industry have occurred, with passive investing gaining momentum. The intersection of these two major trends poses special challenges during market downturns, magnifying portfolio losses and leading to significant outflows. Emerging market (EM) investors have seen two major downturn events in the 2020s, namely the COVID-19 pandemic and the Russia-Ukraine conflict, both of which have strongly affected EM portfolios’ risk-return profiles and increased their correlations with their developed market counterparts, eliminating much or all of EMs’ diversification benefits. This has led to major capital outflows from EM countries, further destabilizing these fragile economies. Against this backdrop, we argue that capital need not exit these riskier markets during periods of turmoil and support this by developing a second-generation Automated Adaptive Trading System (AATS) back-tested on a relevant, diversified EM portfolio that tracks the Morgan Stanley Capital International (MSCI) Emerging Markets Index during a volatile period characterized by negative returns, high risk, and a high correlation with global markets for the buy-and-hold EM portfolio. The system incorporates an Autoregressive Moving Average-Generalized AutoRegressive Conditional Heteroskedasticity model that offers an interpretability advantage over machine-learning methods. The main strength of the AATS is its ability to allow the embedded hybrid forecasting model to adapt to the changing environments that characterize EMs. This is done by implementing a recursive window technique and running a user-specified fitness function to dynamically optimize the mean equation parameters throughout the lead time. Back-testing several configurations of the flexible AATS consistently reveals its superiority while assuring the robustness of the results. We conclude that with the right investment tools, EMs continue to offer compelling opportunities that should not be overlooked. The novel AATS proposed in this study is such a tool, providing active EM investors with substantial value-added through its ability to generate abnormal returns, and can help to enhance the resilience of EMs by mitigating the cost of crises for those countries.
تتأثر الدول النامية بالصراعات الجيوسياسية. في هذا السياق، يهدف الباحث إلى تسليط الضوء على الدور الهام الذي يمكن أن يلعبه تطبيق نموذج الاقتصاد الدائري في تعزيز القدرات التنافسيه و الاقتصاديه لمصر وهيئه قناه السويس وإظهار الدور اللوجستي والمحوري لها، من خلال رفع الكفاءة التشغيليه ودعم سلاسل الامداد المستدامه؛ مما يساهم فى الحفاظ على الموارد الناضبه؛ كما يهدف إلى تقصي تأثير تطبيق نموذج الاقتصاد الدائري على تعزيز القدرات التنافسية والاقتصادية لمصر وهيئة قناة السويس. تأتي أهمية هذا البحث في سياق التحديات الجيوسياسية والاقتصادية العالمية التي تواجه مصر، خاصةً تداعيات الأزمة الروسية والتوترات في منطقة الشرق الأوسط، والتي أثرت سلبًا على سلاسل الإمداد والاستثمار الأجنبي المباشر. وتكمن مشكله البحث فى تعرض الاقتصاد المصري كغيره من العديد من اقتصادات العالم إلى صدمة بداية من عام وتتلخص فى الإجابة على التساؤل حول تأثير تطبيق الاقتصاد الدائري على الكفاءة التشغيلية واستدامة سلاسل الإمداد في هيئة قناة السويس. حيث اعتمد الباحث على المنهج الوصفى التحليلي؛ حيث وصف الحالة الحالية من تطبيق الأقتصاد الدائري في مصر، وقام بتحليل تأثير تطبيق الاقتصاد الدائري في هيئة قناة السويس، واعتمد البحث على المنهجية المختلطه،. تكمن الأهمية العلمية للبحث في كونه دراسة رائدة في مصر، بينما تبرز الأهمية العملية في تقديم رؤى لصناع القرار في هيئة قناة السويس والمساهمة في تحقيق التنمية المستدامة وتخفيف العبء الاقتصادي. تتمثل النتيجة المتوقعة في تقديم "خريطة طريق" لتطبيق الاقتصاد الدائري، تتضمن العوائق، التوصيات، والعوائد المتوقعة، مما يساهم في تحسين الأداء الاقتصادي للقناة ودعم التحول نحو اقتصاد أكثر استدامة في مصر.
Economic theory justifies policy when there are concrete market failures. The article shows how in the case of innovation, successful policies that have led to radical innovations have been more about market shaping and creating through direct and pervasive public financing, rather than market fixing. The paper reviews and discusses evidence for this in three key areas: (i) the presence of finance from public sources across the entire innovation chain; (ii) the concept of ‘mission-oriented’ policies that have created new technological and industrial landscapes; and (iii) the entrepreneurial and lead investor role of public actors, willing and able to take on extreme risks, independent of the business cycle. We further illustrate these three characteristics for the case of clean technology, and discuss how a market-creating and -shaping perspective may be useful for understanding the financing of transformative innovation needed for confronting contemporary societal challenges.
Decentralized Finance (DeFi), a financial ecosystem without centralized controlling organization, has introduced a new paradigm for lending and borrowing. However, its capital efficiency remains constrained by the inability to effectively assess the risk associated with each user/wallet. This paper introduces the 'On-Chain Credit Risk Score (OCCR Score) in DeFi', a probabilistic measure designed to quantify the credit risk associated with a wallet. By analyzing historical real-time on-chain activity as well as predictive scenarios, the OCCR Score may enable DeFi lending protocols to dynamically adjust Loan-to-Value (LTV) ratios and Liquidation Thresholds (LT) based on the risk profile of a wallet. Unlike existing wallet risk scoring models, which rely on heuristic-based evaluations, the OCCR Score offers a more objective and probabilistic approach, aligning closer to traditional credit risk assessment methodologies. This framework can further enhance DeFi's capital efficiency by incentivizing responsible borrowing behavior and optimizing risk-adjusted returns for lenders.
This thesis evaluates most of the extreme mixture models and methods that have appended in the literature and implements them in the context of finance and insurance. The paper also reviews and studies extreme value theory, time series, volatility clustering, and risk measurement methods in detail. Comparing the performance of extreme mixture models and methods on different simulated distributions shows that the method based on kernel density estimation does not have an absolute superior or close to the best performance, especially for the estimation of the extreme upper or lower tail of the distribution. Preprocessing time series data using a generalized autoregressive conditional heteroskedasticity model (GARCH) and applying extreme value mixture models on extracted residuals from GARCH can improve the goodness of fit and the estimation of the tail distribution.
Claudia Biancotti, Carolina Camassa, Andrea Coletta
et al.
Advancements in large language models (LLMs) have renewed concerns about AI alignment - the consistency between human and AI goals and values. As various jurisdictions enact legislation on AI safety, the concept of alignment must be defined and measured across different domains. This paper proposes an experimental framework to assess whether LLMs adhere to ethical and legal standards in the relatively unexplored context of finance. We prompt twelve LLMs to impersonate the CEO of a financial institution and test their willingness to misuse customer assets to repay outstanding corporate debt. Beginning with a baseline configuration, we adjust preferences, incentives and constraints, analyzing the impact of each adjustment with logistic regression. Our findings reveal significant heterogeneity in the baseline propensity for unethical behavior of LLMs. Factors such as risk aversion, profit expectations, and regulatory environment consistently influence misalignment in ways predicted by economic theory, although the magnitude of these effects varies across LLMs. This paper highlights both the benefits and limitations of simulation-based, ex post safety testing. While it can inform financial authorities and institutions aiming to ensure LLM safety, there is a clear trade-off between generality and cost.
Abstract In rural China, the aging population faces unique challenges that affect their life satisfaction. These challenges are compounded by disparities in access to resources compared to urban areas, making it crucial to address these issues to ensure a dignified and fulfilling later life. Understanding and improving life satisfaction for this demographic is essential, not only for the well-being of the elderly but also for the overall social stability and economic sustainability of these communities. Using a rural sample from the Chinese Social Survey (CSS), this study investigates the interplay between social quality factors and life satisfaction among older adults in rural China. It focuses on the roles of socioeconomic security, social cohesion, social inclusion, and social empowerment in shaping the well-being of this demographic. It was found that the four conditional factors of social quality—socioeconomic security, social cohesion, social inclusion, and social empowerment—have different influences on life satisfaction among older adults in rural areas. The statistical model shows that the influence of absolute family income on social and economic security does not have statistical significance. In contrast, housing, pension insurance participation, and public safety perception positively affect life satisfaction among older adults in rural areas. Among the social cohesion factors, higher social morality, legal system evaluation, social identity, grassroots government, and interpersonal trust contribute to the life satisfaction of older adults in rural areas. Regarding social inclusion factors, good social tolerance, social equity, and perception of government public services can significantly improve life satisfaction among older adults in rural areas. For social empowerment factors, social participation helps expand the social support network of older adults in rural areas, enhancing their life satisfaction. Paths to improving the quality of rural society should be explored to improve relative poverty in rural areas. They should continue to be pursued to strengthen the old-age security system for older adults in rural areas. Further, the National Medium- and Long-Term Plan for Actively Responding to Population Aging should be the base for forming an elderly-friendly society. A good atmosphere promotes the participation of people, families, and society; supports social cohesion; enhances inclusion; and promotes social participation, improving life satisfaction.
Mustafa Tevfik Kartal, Serpil Kılıç Depren, Ugur Korkut Pata
et al.
Abstract This study constructs a proposed model to investigate the link between environmental, social, and governance (ESG) disclosures and ESG scores for publicly traded companies in the Borsa Istanbul Sustainability (XUSRD) index. In this context, this study considers 66 companies, examining recently structured ESG disclosures for 2022 that were published for the first time as novel data and applying a multilayer perceptron (MLP) artificial neural network algorithm. The relevant results are fourfold. (1) The MLP algorithm has explanatory power (i.e., R2) of 79% in estimating companies’ ESG scores. (2) Common, environment, social, and governance pillars have respective weights of 21.04%, 44.87%, 30.34%, and 3.74% in total ESG scores. (3) The absolute and relative significance of each ESG reporting principle for companies’ ESG scores varies. (4) According to absolute and relative significance, the most effective ESG principle is the common principle, followed by social and environmental principles, whereas governance principles have less significance. Overall, the results demonstrate that applying a linear approach to complete deficient ESG disclosures is inefficient for increasing companies’ ESG scores; instead, companies should focus on the ESG principles that have the highest relative significance. The findings of this study contribute to the literature by defining the most significant ESG principles for stimulating the ESG scores of companies in the XUSRD index.
In this article, the author analyzes the market of debt instruments in the field of «green» financing on the example of «green» bonds issued by the city of Moscow. The placement of “green” bonds is targeted, the funds raised should be used to finance environmental projects. The Moscow government issued green bonds totaling 72 billion rubles to finance the replacement of the bus fleet and the construction of metro stations, which will reduce the emission of pollutants and greenhouse gases into the atmosphere. Consequently, investors, by purchasing “green” bonds, not only successfully invest free cash, but also finance an investment project, the purpose of which is to improve the climatic conditions of the city of Moscow. The city of Moscow is an issuer with high credit ratings, which is achieved due to the budget balance and low debt burden. The issue of green bonds of Moscow is the first case of placement of sub-federal securities of green financing. From this point of view, they are of interest both for investors wishing to invest in environmental projects, and for government agencies and other subjects of Russia. To distribute this type of financial instrument, the efforts of the state, financial institutions and corporate investors should be combined. Measures taken by the Government of the Russian Federation include certification of the taxonomy process for environmental projects and the adoption of Russian standards for the development of sustainable development bonds, including public sector green bonds. Government support measures will help develop infrastructure and increase investor confidence in green financial instruments.