Nikeethan Selvaratnam, Dorinel Bastide, Clément Fernandes, Wojciech Pieczynski
Apr 23, 2026·q-fin.RM·PDF Predicting future operational risk losses gives rise to a significant challenge due to the heterogeneous and time-dependent structures present in real-world data. Furthermore, stress test exercises require examining the relationship with operational losses. To capture such relationship, we propose to use an extension of Hidden Markov Models to multivariate observations. This model introduces a third auxiliary variable designed to accommodate the economic covariates in the time-series data. We detail the unique aspects of operational risk data and describe how model calibration is achieved via the Expectation-Maximization (EM) algorithm. Additionally, we provide the calibration results for the various risk-event types and analyze the relevance of the inclusion of the macroeconomic covariates.
Lennart Ante, Friedrich-Philipp Wazinski, Aman Saggu
Apr 23, 2026·q-fin.GN·PDF Biodiversity loss is accelerating at an unprecedented pace, threatening ecosystem stability, economic resilience, and human well-being, with billions required to reverse current trends. Against this backdrop, biodiversity finance has emerged as a rapidly expanding but highly fragmented field spanning ecology, economics, finance, accounting, and policy. However, it remains emerging and complex, with the majority of relevant knowledge being produced in non-finance journals. This study employs quantitative bibliometric analysis to examine a corpus of 189,456 references underlying 3,998 articles related to biodiversity and finance. The analysis identifies eight primary research streams within the field that concern (1) strategic and financial approaches in global biodiversity conservation, (2) the impact and implementation of payments for environmental services (PES) in developing countries, (3) neoliberal influences and implications in environmental conservation, (4) biodiversity offsets and conservation, (5) ecosystem services and biodiversity, (6) integrating conservation and community interests in biodiversity management, (7) balancing agricultural intensification with biodiversity conservation, and (8) global and corporate biodiversity reporting. The characteristics of each research stream and its prevalent publications are outlined, alongside an analysis of their temporal evolution and the degree of information exchange among the research streams. The findings provide a structured map of the intellectual architecture of biodiversity finance, document pronounced silos between economically-oriented and critical/political-economy research streams, and translate these patterns into a focused research agenda and implications for policymakers, financial institutions, and corporate actors.
Sebastian Lehner, Alejandro Lopez-Lira
Apr 23, 2026·q-fin.GN·PDF Frozen large language model (LLM) checkpoints extract information from pre-cutoff public text that is associated with future fundamentals and equity returns beyond standard contemporaneous valuation measures. Because each frozen checkpoint has a fixed knowledge cutoff, it can be interpreted as a compressed representation of publicly available textual information at a given point in time. We treat twelve OpenAI snapshots spanning 2021-2025 as time-stamped summaries of the public textual record and extract a sector-neutral LLM outlook score for roughly 7,000 U.S. equities per cross-section. The outlook score is positively associated with analyst revisions, target-price changes and one-month cross-sectional returns in both Fama-MacBeth regressions and pooled panels with model fixed effects (t = 6.02), after direct controls for market-implied valuation and standard factors. Predictability broadly increases with the return horizon, despite a non-monotonic intermediate dip, and, in the pooled panel, is stronger for firms with high analyst coverage, consistent with the view that the bottleneck is not investor inattention but the cost of aggregating dispersed qualitative information across many documents.
Useong Shin
Apr 21, 2026·q-fin.GN·PDF Put-call parity holds under risk-neutral pricing, yet enforcement exposes arbitrageurs to path-dependent capital costs. The carry gap-the annualized wedge between option-implied and OIS discount factors-is a Q-measure object, but P-measure investment opportunities may shape its enforcement burden. We document this alignment in SPX and RUT options: low-frequency global asset returns raise in-sample R^2 by 0.093 and 0.082 and lift pooled out-of-sample R^2 from 0.221 to 0.364 (SPX) and 0.171 to 0.309 (RUT). Effective horizons differ by asset-IEFA (70 days), IGOV (400 days), IAU (336 days)-and asset terms largely absorb the OIS baseline, providing systematic evidence of a P-Q channel.
Useong Shin
Apr 21, 2026·q-fin.GN·PDF Put-call parity is a terminal-payoff identity; quoted residuals against traded futures are near zero. Yet enforcing parity is path-dependent, exposing arbitrageurs to daily settlement, margin, and finite capital. Using minute-level NBBO data on S&P 500 and Russell 2000 options, I extract option-implied discount factors, compare them with the OIS curve, and construct an annualized carry gap. A reduced-form specification centered on a volatility times sqrt(tau) path-risk term links the carry gap to implementation risk, trading frictions, and financial conditions, with coefficient signs stable across leave-one-year-out validation. The carry gap is an implementation wedge invisible in price space but systematic in carry space.
Yikuan Huang, Zheqi Fan, Kaiqi Hu, Yifan Ye
Apr 21, 2026·q-fin.PM·PDF Text-based financial networks are increasingly used to study cross-stock return predictability. A common approach constructs links from similarities in firms' disclosure embeddings, but such networks often contain spurious edges because textual proximity does not necessarily imply economic connection. We propose a two-stage framework that first builds a sparse candidate graph from 10-K embeddings and then uses a large language model to classify and filter candidate edges according to their economic relations. The refined graph is used to aggregate pair-level mean-reversion signals into stock-level trading signals with relation-aware and distance-based weights. In a backtest on S&P 500 constituents from 2011 to 2019, LLM-based edge filtering improves the long-short Sharpe ratio from 0.742 to 0.820 and reduces maximum drawdown from $-$10.47% to $-$7.85%. These results suggest that LLM-based reasoning can improve the economic fidelity of text-derived financial networks and strengthen cross-stock predictability.
Terence Lim, Kumar Muthuraman, Michael Sury
We introduce a multi-agent framework intended to emulate parts of a quantitative research team and support equity factor research on large financial panel datasets. QRAFTI integrates a research toolkit for panel data with MCP servers that expose data access, factor construction, and custom coding operations as callable tools. It can help replicate established factors, formulate and test new signals, and generate standardized research reports accompanied by narrative analysis and computational traces. On multi-step empirical tasks, using chained tool calls and reflection-based planning may offer better performance and explainability than dynamic code generation alone.
Shumiao Ouyang, Pengfei Sui
We study how AI agents form expectations and trade in experimental asset markets. Using a simulated open-call auction populated by autonomous Large Language Model (LLM) agents, we document three main findings. First, AI agents exhibit classic behavioral patterns: a pronounced disposition effect and recency-weighted extrapolative beliefs. Second, these individual-level patterns aggregate into equilibrium dynamics that replicate classic experimental findings (Smith et al., 1988), including the predictive power of excess demand for future prices and the positive relationship between disagreement and trading volume. Third, by analyzing the agents' reasoning text through a twenty-mechanism scoring framework, we show that targeted prompt interventions causally amplify or suppress specific behavioral mechanisms, significantly altering the magnitude of market bubbles.
Nima Afsharhajari, Jonathan Yu-Meng Li
Apr 18, 2026·q-fin.GN·PDF Sparsity or complexity? In modern high-dimensional asset pricing, these are often viewed as competing principles: richer feature spaces appear to favor complexity, while economic intuition has long favored parsimony. We show that this tension is misplaced. We distinguish capacity sparsity-the dimensionality of the candidate feature space-from factor sparsity-the parsimonious structure of priced risks-and argue that the two are complements: expanding capacity enables the discovery of factor sparsity. Revisiting the benchmark empirical design of Didisheim et al. (2025) and pushing it to higher complexity regimes, we show that nonlinear feature expansions combined with basis pursuit yield portfolios whose out-of-sample performance dominates ridgeless benchmarks beyond a critical complexity threshold. The evidence shows that the gains from complexity arise not from retaining more factors, but from enlarging the space from which a sparse structure of priced risks can be identified. The virtue of complexity in asset pricing operates through factor sparsity.
Andrew Y. Chen
Apr 18, 2026·q-fin.GN·PDF AI stocks trade at extraordinary valuations. We develop an asset pricing model in which investors use AI stocks to hedge against an AI singularity that displaces their consumption. Because markets are incomplete -- investors cannot trade private AI capital -- AI stocks command a premium. Market incompleteness distorts both valuations and the efficient development of AI, creating a rationale for government transfers that becomes compelling when singularity-driven growth overwhelms deadweight costs. This paper was generated by AI, using https://github.com/chenandrewy/ralph-wiggum-asset-pricing/.
Songrun He
Apr 15, 2026·q-fin.GN·PDF In this paper, I present the first comprehensive, around-the-clock analysis of systematic jump risk by combining high-frequency market data with contemporaneous news narratives identified as the underlying causes of market jumps. These narratives are retrieved and classified using a state-of-the-art open-source reasoning LLM. Decomposing market risk into interpretable jump categories reveals significant heterogeneity in risk premia, with macroeconomic news commanding the largest and most persistent premium. Leveraging this insight, I construct an annually rebalanced real-time Fama-MacBeth factor-mimicking portfolio that isolates the most strongly priced jump risk, achieving a high out-of-sample Sharpe ratio and delivering significant alphas relative to standard factor models. The results highlight the value of around-the-clock analysis and LLM-based narrative understanding for identifying and managing priced risks in real time.
Hilde C. Bjornland, Nicolas Hardy, Dimitris Korobilis
We develop a Quantile Bayesian Vector Autoregression (QBVAR) to forecast real oil prices across different quantiles of the conditional distribution. The model allows predictor effects to vary across quantiles, capturing asymmetries that standard mean-focused approaches miss. Using monthly data from 1975 to 2025, we document three findings. First, the QBVAR improves median forecasts by 2-5\% relative to Bayesian VARs, demonstrating that quantile-specific dynamics matter even for point prediction. Second, uncertainty and financial condition variables strongly predict downside risk, with left-tail forecast improvements of 10-25\% that intensify during crisis episodes. Third, right-tail forecasting remains difficult; stochastic volatility models dominate for upside risk, though forecast combinations that include the QBVAR recover these losses. The results show that modeling the conditional distribution yields substantial gains for tail risk assessment, particularly during major oil market disruptions.
Hanming Fang, Xian Gu, Hanyin Yan, Wu Zhu
We develop a high-precision classifier to measure artificial intelligence (AI) patents by fine-tuning PatentSBERTa on manually labeled data from the USPTO's AI Patent Dataset. Our classifier substantially improves the existing USPTO approach, achieving 97.0% precision, 91.3% recall, and a 94.0% F1 score, and it generalizes well to Chinese patents based on citation and lexical validation. Applying it to granted U.S. patents (1976-2023) and Chinese patents (2010-2023), we document rapid growth in AI patenting in both countries and broad convergence in AI patenting intensity and subfield composition, even as China surpasses the United States in recent annual patent counts. The organization of AI innovation nevertheless differs sharply: U.S. AI patenting is concentrated among large private incumbents and established hubs, whereas Chinese AI patenting is more geographically diffuse and institutionally diverse, with larger roles for universities and state-owned enterprises. For listed firms, AI patents command a robust market-value premium in both countries. Cross-border citations show continued technological interdependence rather than decoupling, with Chinese AI inventors relying more heavily on U.S. frontier knowledge than vice versa.
Benjamin Leroy, Davi Marim, El Ghali Benjelloun, Arthur Rozan Debeaurain, Jean-Michel Dalle
We explore a quantitative approach to emerging technological sovereignty and geoeconomic power by assessing the relative positioning of countries with economic complexity methods applied to the structure of national venture-capital (VC) portfolios and their associated Revealed Venture Advantage (RVA) metrics. Using Crunchbase firm- and deal-level data, we map venture-backed startups to 18 emerging technology domains via a probabilistic multi-label large-language-model classifier, and construct an RVA-based country-technology specialization matrix for the 17 countries with the highest aggregate VC funding. From this matrix, we derive two eigenvector-based measures: a Geoeconomic Complexity Index (GCI) that ranks countries by the composition of their venture specializations, and an Emerging Technology Geoeconomic Complexity Index (ETGCI) that ranks domains by the extent to which specialization is concentrated among high-GCI countries. Empirically, Cloud Computing, Cybersecurity Tools, and Medtech exhibit the highest ETGCI values, reflecting concentration of specialization in a small set of leading countries. The United States and Israel consistently occupy a marked "high-diversity/low-ubiquity" position and lead the GCI ranking, followed by China, France, Japan, and Germany; both country and domain rankings are stable from 2021-2024. Finally, relatedness-based simulations identify, when it exists, for each country the Simplest Single Sovereignty Enhancing Technology (SSSET), i.e., the most feasible single new technological direction associated with the largest expected improvement in relative geoeconomic positioning.
Grace Burtenshaw, Ashley Burtenshaw, Meagan Carney
Apr 10, 2026·q-fin.GN·PDF In recent years Australia has observed a growing, unexplained resilience of increasing house price trends. Here, we seek to understand what is driving Australia's indestructible asset using insights from market experts. We construct a differential equation model of house price to develop intuition for its historical behaviour and responsiveness to changes in mortgage rates. Using this model, we identify a point of 'decoupling' between house price and mortgage rate in the system with supply limitations found to be the main driver for this change. From there, modern extreme value techniques are implemented on real-world data to investigate how the effectiveness of mortgage rate in moderating extreme house price has changed before and after this historical decoupling. We find that without an increase in the housing supply chain, through either deregulation or reduced competition with government building, an 11\% increase in mortgage rate will be needed to slow extreme housing costs.
Mykola Babiak, Jozef Barunik, Josef Kurka
Cross-sectional dispersion in firm-level realized skewness is significantly and negatively related to future stock market returns. The predictive power of skewness dispersion is robust to in-sample and out-of-sample estimation and is incremental over a broad set of existing predictors, with only a few alternatives retaining independent explanatory ability. Skewness dispersion also delivers substantial economic gains in portfolio allocation. Its forecasting power is concentrated in months with monetary policy announcements, reflecting an information-based mechanism. The empirical evidence suggests that skewness dispersion captures the gradual incorporation of macro news into prices, which is driven by variation in aggregate risk and valuation adjustments.
Junliang Luo, Xihan Xiong, Zonglun Li, Hong Kang, Xue Liu, William J Knottenbelt, Katrin Tinn
The global financial architecture is undergoing a shift from intermediary centric-settlement to programmable infrastructure, to transmute trillions in static illiquid capital into active, high-velocity instruments. We argue that Real World Asset (RWA) tokenization represents a conceptual evolution beyond mere digitization, converting passive ledger entries into programmable economic agents capable of autonomous settlement and algorithmic collateralization. However, achieving such seamless capital efficiency necessitates resolving the fundamental friction between deterministic on-chain code and probabilistic off-chain reality, navigating the oracle problem and jurisdictional interoperability. This systematization of knowledge presents a taxonomy for the RWA lifecycle and deconstructs the multi-layered architecture, spanning legal custody, technical standards, and cryptoeconomic valuation, required to enforce off-chain rights within on-chain environments. We study systemic constraints such as latency and regulatory fragmentation through a comparative overview of sovereign debt, private credit, and real estate protocols, complemented by an empirical case study of on-chain U.S. Treasuries. We synthesize these findings to propose a prognostic outlook, positing that while asset tokenization provides a transitional bridge, it is not necessarily the inevitable shift compared to the emergence of unified, programmable ledgers.
Yuanzhe Zhang, Yuexin Xiang, Yuchen Lei, Qin Wang, Tian Qiu, Yujing Sun, Spiridon Zarkov, Tsz Hon Yuen, Andreas Deppeler, Jiangshan Yu, Kwok-Yan Lam
Agentic AI rivals human capabilities across a wide range of domains. Looking ahead, it is foreseeable that AI agents will autonomously handle complex workflows and interactions. Early prototypes of this paradigm are emerging, e.g., OpenClaw and Moltbook, signaling a shift toward Agent-to-Agent (A2A) ecosystems. However, despite these promising blueprints, critical trust and security challenges remain, particularly in scenarios involving financial transactions. Ensuring secure and reliable payment mechanisms between unknown and untrusted agents is crucial to complete a fully functional and trustworthy A2A ecosystem. Although blockchain-based infrastructures provide a natural foundation for this setting, via programmable settlement, transparent accounting, and open interoperability, trust and security challenges have not yet been fully addressed. Hence, for the first time, we systematize blockchain-based A2A payments, e.g., X402, with a four-stage lifecycle: discovery, authorization, execution, and accounting. We categorize representative designs at each stage and identify key challenges, including weak intent binding, misuse under valid authorization, payment-service decoupling, and limited accountability. We highlight future directions for strengthening cross-stage consistency, enabling behavior-aware control, and supporting compositional payment workflows across agents and systems.
Zhenyu Gao, Wenxi Jiang, Yutong Yan
Prior research shows that large language models (LLMs) exhibit systematic extrapolation bias when forming predictions from both experimental and real-world data, and that prompt-based approaches appear limited in alleviating this bias. We propose a supervised fine-tuning (SFT) approach that uses Low-Rank Adaptation (LoRA) to train off-the-shelf LLMs on instruction datasets constructed from rational benchmark forecasts. By intervening at the parameter level, SFT changes how LLMs map observed information into forecasts and thereby mitigates extrapolation bias. We evaluate the fine-tuned model in two settings: controlled forecasting experiments and cross-sectional stock return prediction. In both settings, fine-tuning corrects the extrapolative bias out-of-sample, establishing a low-cost and generalizable method for debiasing LLMs.
Andrew Ang, Nazym Azimbayev, Andrey Kim
Agentic AI shifts the investor's role from analytical execution to oversight. We present an agentic strategic asset allocation pipeline in which approximately 50 specialized agents produce capital market assumptions, construct portfolios using over 20 competing methods, and critique and vote on each other's output. A researcher agent proposes new portfolio construction methods not yet represented, and a meta-agent compares past forecasts against realized returns and rewrites agent code and prompts to improve future performance. The entire pipeline is governed by the Investment Policy Statement--the same document that guides human portfolio managers can now constrain and direct autonomous agents.