This paper investigates whether machine learning forecasts of hourly BTC-USDT returns can be converted into economically meaningful trading performance after transaction costs. Using approximately 70,000 hourly observations from 2018-2026, XGBoost, LSTM, and iTransformer are evaluated in a 27-fold walk-forward protocol. All three models produce positive gross trading performance in selected configurations, but naive sign-based strategies fail once transaction costs of ten basis points are imposed. A cost-aware execution filter, which prevents trades only when the forecast magnitude exceeds a transaction-cost-based threshold, sharply reduces turnover and restores profitability in selected configurations. The strongest long-only XGBoost strategy produces annualised returns above 65% with a Sharpe ratio above one. Additional tests show that technical indicators improve performance in selected cases, EGARCH-derived features do not provide uniformly robust gains, and XGBoost is descriptively stronger than the neural alternatives, although bootstrap evidence does not support formal statistical dominance. Loss-function and model-selection effects are secondary and statistically fragile. The results show that the main obstacle in hourly cryptocurrency trading is not only weak predictability, but also the way forecasts are converted into trades.

We study how frontier large language models (LLMs) behave as financial forecasters during boom-bust market cycles when made progressively aware of Soros's theory of reflexivity. Standard AI-assisted forecasting treats the market as an exogenous system. Reflexivity theory holds otherwise: prices shape fundamentals, and every forecaster is a participative agent in the loop it analyzes. We evaluate three frontier models - GPT5, Claude Sonnet 4.6, and Gemini 3 Pro - under four accumulating zero-shot conditions across two historically distinct episodes: the dot-com bubble (1996-2001) and the global financial crisis (2004-2009). The primary metric is directional forecasting accuracy; we also report the Sharpe ratio of an implied long/cash strategy to capture the risk-adjusted economic value of the forecasts. All inputs are anonymized and normalized to guard against memorization. We find that conditions incorporating reflexivity awareness improve forecasting accuracy differently across models and context windows, revealing that the same theoretical awareness can produce qualitatively different forecasting behavior across frontier LLMs.

Bitcoin price prediction has attracted hundreds of academic papers and continuous social media debate, yet the field lacks consensus on even basic questions: can any model beat a naive "today's price" baseline at horizons of one to six months? We survey the peer-reviewed landscape, categorize papers by evaluation methodology, and contrast academic findings with informal but substantive discourse on X/Twitter. The picture that emerges is sobering. At short-to-medium horizons, no peer-reviewed study has shown robust superiority over the naive baseline across multiple market regimes. Daily predictability is real but does not extend to hourly or monthly horizons, and may not survive transaction costs. The stock-to-flow model has failed formal out-of-sample testing, and Metcalfe's Law valuations have been challenged as spurious. The Bitcoin price power law, while empirically compelling, has not been subjected to formal distributional tests. Meanwhile, social media practitioners raise valid statistical critiques -- ordinary least squares (OLS) violations, backtest overfitting, spurious regressions -- that the academic literature has not formalized. We identify open research directions and propose concrete methodological standards for future work -- walk-forward evaluation, multi-regime holdout windows, naive baseline comparison, inclusion of zero in hyperparameter grids, and Diebold-Mariano significance testing -- arguing that the field's primary need is not more models but better evaluation.

Financial markets are inherently non-stationary, exhibiting frequent regime shifts and structural changes that render traditional Portfolio Management (PM) approaches ineffective. Existing remedies, such as rolling-window retraining and naive online fine-tuning, are hindered by high computational costs and insufficient knowledge utilization, respectively, resulting in low returns and limited adaptability. Continual learning (CL) offers a promising paradigm by enabling trading agents to accumulate and transfer knowledge across sequential tasks. In this paper, we propose \textbf{Re}gime-aware \textbf{C}ontinual \textbf{A}daptive \textbf{P}ortfolio management (\textbf{ReCAP}), a novel framework that integrates CL into PM to address the challenges of dynamic financial environments. ReCAP employs an adaptive regime detection module to segment historical market data into variable-length regimes, enabling regime-specific learning of policy vectors and the construction of a policy library. During continual trading, a regime-gate module adaptively combines policy vectors from the library based on the current market state, facilitating rapid adaptation to newly detected regimes. Only the regime-gate and the current regime's policy vector are continually updated to preserve useful knowledge effectively. Extensive experiments on five real-world datasets demonstrate that ReCAP consistently outperforms popular baselines, achieving superior returns in long-term investment horizons and rapid adaptation to regime shifts.

The damage to transportation infrastructure caused by disasters can indirectly lead to economic damage through economic interdependence, even in areas that are not directly affected. However, even when transportation routes are interrupted by a disaster, the damage can be mitigated if alternative routes are secured. Rural areas with low-density transportation networks are more vulnerable to traffic disruptions in a disaster. This study develops a method for evaluating the effectiveness of redundant transportation networks in mitigating economic vulnerability in the event of a disaster. Our methodology combines inter-regional road network connectivity with a spatial computable general equilibrium (SCGE) model. We apply the method to road disruption scenarios in the Chugoku region of Japan, which has a system of parallel highways. The affected areas are in close geographical proximity to many rural areas and have strong economic interdependencies with them. Several counterfactual simulations depicted the situation without the alternative road and the disaster. We evaluate the transportation impacts, measured by changes in travel time, and the economic impacts, measured by negative benefits, respectively. The results suggest that the economic vulnerability reduction effect is more far-reaching than the transportation impacts.

Deep learning models show promise in financial forecasting, yet their generalization is often undermined by small datasets, noisy signals, and non-stationarity. While meta-learning and related techniques mitigate some of these issues, they typically do not account for a core limitation in macro-financial prediction: the scarcity of distinct macroeconomic regimes that drive asset returns. We introduce HANET (Hierarchical Attention Network), a hybrid LSTM-based architecture that integrates macroeconomic domain knowledge through attention over long-run macro contexts while preserving high-frequency market dynamics. HANET organizes information in a hierarchical mixed-frequency structure, with daily asset-return signals nested within monthly macroeconomic windows, and introduces a Hierarchical Cross-Attention mechanism that reconciles low-frequency macro signals with high-frequency returns without discarding granular daily information. By framing regime selection as attention over macroeconomic contexts, the model adapts to scarce and shifting regimes. Empirically, across 55 liquid futures spanning multiple asset classes, HANET consistently outperforms neural forecasters that ignore macroeconomic information, particularly during turbulent periods, improving risk-adjusted returns and mitigating losses. Ablation studies show that these gains rely on structured macro conditioning rather than naive feature augmentation: an LSTM with the same macro representation performs poorly, and shuffling macro contexts substantially degrades performance. Finally, HANET provides interpretability through attention weights, highlighting which historical regimes are most influential for each forecast and linking macro conditions to portfolio outcomes. These results establish HANET as a systematic approach to integrating macroeconomic information into attention-based deep learning for financial forecasting.

Q-variance (so-called) posits a statistical relationship $\mathbf{E}(\sigma^2 | z) = \sigma_0^2 + \tfrac{1}{2}z^2$ between an asset's volatility $\sigma^2$, as observed in a time interval $T$, and its (suitably scaled) return $z$ in the same interval. We here show that this relationship is {\em exactly equivalent} to to positing an Inverse Gamma probability distribution for $\sigma^2$ itself. We then show that such a distribution is exactly generated by a multiplicative Langevin process with an arbitrary, settable coherence time $\tau_c$, so that very nearly the same Q-variance relationship will hold for all $T \ll \tau_c$.

Recession indicators are often viewed as U.S. specific, raising the question of whether labor market-based rules such as the Sahm Rule and the Michez Rule can reliably detect recessions in other countries. To answer this, we evaluate whether such rules can be adapted to Japan by calibrating thresholds and smoothing parameters to Japanese labor market data. We construct a large set of 95,832 recession indicators combining unemployment and vacancy data. The selected classifiers are statistically perfect as they identify all 11 historical recessions in the 1970-2021 training period without generating any false positives. Among these, 193 classifiers lie on the anticipation-precision frontier. Restricting attention to the high-precision segment yields six classifiers with a standard deviation of detection errors below 3 months. The selected classifier ensemble signals recessions, on average, 0.06 months after their true onset. Overall, these findings suggest that slack-based labor market rules provide a general framework for improving real-time recession detection across countries.

This paper examines whether real-time GDP announcements can reliably identify business-cycle turning points. Using U.S. real-time GDP vintages from 1947 to 2021, we construct 4,356 recession indicators based on alternative smoothing methods and scaling variations. We then combine these indicators with alternative thresholds to generate 137,457 perfect recession classifiers. The selected classifiers identify all 12 historical recessions without generating false positives or false negatives. Restricting attention to the high-precision segment yields two classifiers with a standard deviation of detection errors below three months, while the selected ensemble signals recessions, on average, 3.04 months after their official onset. The framework accurately identifies recession episodes across vintages, suggesting that discrepancies in prior work may reflect limitations of traditional dating methods in addition to data revisions. Overall, the results indicate that real-time GDP announcements provide a practical proxy for NBER-style recession dating.

The GDP of a country is modelled as the relative interaction between two agents - working hours, reflecting the social choice of a population, and Total Factor Productivity, reflecting the collective investment in productivity enhancers. It is shown that a Random Forest model can accu- rately predict the GDP from these two factors. The differences in the choices made by Germany and USA are analysed though Gini importance, SHAP plots and partial dependency. It is shown that the differences in the social structure of the countries are reflected in the relative contribution of working hours and productivity to the GDP.

This paper analyzes transaction fees on blockchains by considering that they form a priority queue and users play a queueing game. Using an M/G^K/1 priority queue model, we provide new insights into the dynamics governing transaction fees and their impact on user behavior. We derive semi-closed form expressions for steady-state quantities and extend the relationship between user delay costs and transaction fees to general block generation times. We apply the model to the Bitcoin network and simulate user responses under various scenarios. Cross-chain analysis across Bitcoin, Dogecoin, and Litecoin reveals similarities in normalized cost structures.

We introduce a novel real-time dataset, Companies House Real-Time (CHRT), that captures daily firm creation and dissolution activity for the full population of UK-registered companies. CHRT provides a timely measure of business formation, becoming available months before official business demography statistics. We show that incorporation activity leads taxable business births and contains forward-looking information about employment and output growth. Consistent with this, a structural vector autoregression (SVAR) indicates that positive shocks to firm entry generate persistent increases in employment and output.

We describe a library of mathematical finance built in the Lean 4 proof assistant, on top of Mathlib and the BrownianMotion package. It is broad: more than two hundred sorry-free theorems across eleven areas, from the measure-theoretic foundations of continuous-time stochastic calculus through derivative pricing to applied risk, portfolio, and fixed-income theory, and, to our knowledge, the most comprehensive machine-checked development of mathematical finance to date. Breadth is the setting, not the point. Two things make it more than a catalogue. It reaches into the continuous theory far enough to construct the L2 Itô integral as a bounded linear isometry and to derive, rather than assume, the risk-neutral pricing measure. And it audits its own faithfulness: every result is classified by how its Lean statement relates to the mathematics it claims, and a build-enforced gate pins the axioms each proof actually uses, so a reader can see precisely what has been proved and what has only been proved under added hypotheses. We close with a candid finding: a formal base over classical financial mathematics yields certified unification of known results rather than new financial theory. The contribution is therefore methodological and infrastructural, reusable verified foundations for mathematical finance, together with the faithfulness audit.

In inventory market making, the running-penalty coefficient $\phi$ of the Cartea-Jaimungal framework and the risk-aversion parameter $\gamma$ of the Avellaneda-Stoikov framework are typically treated as independent free parameters, calibrated separately. We show that they are in fact not independent. A small set of axioms on the market maker's dynamic preference functional, namely cash-additivity, normalization, concavity, strong dynamic consistency, and law-invariance, forces the preference functional to be the entropic certainty-equivalent on liquidation-adjusted terminal wealth, parametrized by a single positive scalar $\gamma$. The Avellaneda-Stoikov framework is the unique representative of this axiom class. The Cartea-Jaimungal framework is its second-order Taylor expansion in inventory magnitude, with the running coefficient forced to $\phi = \gamma\sigma^2/2$ and (under a mild regularity condition on the liquidation cost) the terminal coefficient forced to $\alpha = \frac{1}{2}L''(0)$. The two frameworks, typically presented as competing alternatives with the choice between them driven by tractability, are different manifestations of a single underlying object. The forced relation is invertible, $\gamma = 2\phi/\sigma^2$, giving a consistency cross-check on independently calibrated desk parameters.

The rapid expansion of artificial intelligence (AI) investment has revived a recurrent question in financial economics: are AI-related assets experiencing a bubble, or is the market capitaliz- ing a durable general-purpose technology? This paper develops a hybrid review and diagnostic framework for evaluating whether AI is in an ongoing financial bubble as of May 2026. The analysis begins from asset-pricing foundations in state prices, stochastic discount factors, martingale valuation, and pricing kernels, then connects these foundations to rational bubbles, behavioral bubbles, technology manias, and modern econometric bubble-detection methods. Current evidence shows both genuine fundamentals and bubble-like fragilities. On the fundamental side, realized revenue growth, enterprise adoption, and productivity evidence support a nontrivial share of AI valuations. On the fragile side, capital expenditure has accelerated faster than observed monetization in some layers, private- market valuations are concentrated in a small number of firms, and investor narratives often capitalize future productivity gains before they have appeared in cash flows. The paper proposes a five-pillar diagnostic framework that combines fundamental valuation, residual-exuberance tests, SADF/GSADF explosive-root procedures, LPPL/HLPPL price-pattern diagnostics, sen- timent and issuance measures, and capex-payback analysis. The central conclusion is that AI is best understood as a real technological revolution with localized bubble dynamics rather than as either a pure speculative mania or a bubble-free productivity miracle.

We consider the problem of estimating the true Sharpe ratio of an asset selected for having the highest observed in-sample Sharpe ratio among many assets. We discuss estimators based on the polyhedral lemma, James Stein shrinkage, debiasing the expected maximum Sharpe ratio, thresholding and empirical Bayes. We test these estimators in simulations, computing bias and root mean square error across different values of sample size, number of assets, and spread and shape of population Sharpe ratios. We also compute rank correlation of the estimators against the underlying quantity, simulating how these estimators might be used to compare or rank the output of different teams which perform this selection process. We find that the James Stein estimator provides the best performance across many different realistic values of the relevant parameters, followed by the GMLEB estimator of Jiang and Zhang. These results are fairly robust to correlation of asset returns, with some caveats.

We present a study of the leading-order asymptotics for VIX option prices in Bergomi models in the short-maturity and small volatility-of-volatility regimes. Both out-of-the-money (OTM) and at-the-money (ATM) asymptotics are considered for one-factor, two-factor Bergomi and $N$-factor models. The leading-order asymptotics are obtained in closed-form, which are translated into predictions for the small-maturity asymptotics of the VIX implied volatility. Numerical illustrations are provided to illustrate the efficiency of the closed-form asymptotic formulas.

This paper investigates the conditions under which the Easterlin hypothesis holds within a neoclassical overlapping generations model with endogenous capital accumulation, wages, interest rates, and fertility. We develop a tractable analytical framework that maps economic transitions into utility space via a continuously differentiable first-order difference equation for cohort lifetime utilities. This reformulation allows for a transparent normative evaluation of non-steady-state paths without requiring explicit solutions to the underlying nonlinear system. Within this framework, we show that when fertility cycles emerge and children are normal goods, the utility of small cohorts strictly exceeds that of large cohorts. Crucially, this cohort-welfare asymmetry is driven by fertility preferences and is independent of the economy's position relative to the golden rule.

This paper studies how workers' beliefs about pay shape the tradeoffs between pay and workplace amenities. We design a multi-stage incentivized survey experiment that combines hypothetical choice experiments with elicited beliefs about starting salaries in real jobs and randomly varies the provision of explicit pay information. Although stated preferences imply sizable willingness to pay for amenities consistent with prior literature, baseline beliefs about salaries in real jobs are systematically biased along two margins: respondents under-predict starting salaries by 18% and expect higher-amenity jobs to pay more, substantially over-predicting the amenity-pay gradient. Exposure to pay information raises mean pay beliefs for similar jobs by 4% and reduces belief dispersion by 15%, but does not alter the strong positive association between perceived pay and advertised amenities, leaving the amenity-pay tradeoffs in stated choices essentially unchanged. While workers have strong preferences for workplace amenities, the tradeoffs they perceive deviate sharply from those present under full information.

Large language models now power robo-advisors and trading agents, yet whether they carry built-in biases toward specific assets is largely untested. We ask three questions: do LLMs systematically prefer certain financial instruments; can an internal representation with causal leverage over those preferences be identified; and does that representation affect downstream financial decisions? We develop a three-level audit protocol and apply it to Bitcoin. First, a behavioral audit of eight frontier LLMs shows that Bitcoin's ranking among money-like instruments is frame-dependent: models place it around rank 5 of 8 as "reliable money" but near the top under crisis and autonomous-agent frames, and an attribute-swap experiment confirms rankings track functional properties, not names. Second, we open a model's internals: a search across thousands of sparse-autoencoder features in Gemma 3 identifies a dominant Bitcoin-selective feature. Amplifying it shifts the model toward the asset and suppressing it shifts the model away, even when "Bitcoin" never appears in the prompt. Third, we test financial consequences: amplification raises Bitcoin's portfolio share by 5.2 percentage points while suppression lowers it by 4.6 pp, with amplification reallocating within crypto and suppression cutting total crypto exposure. We characterize this as bounded behavioral leverage (leverage meaning causal influence over outputs, not financial leverage): an identifiable internal feature can be perturbed to move financial choices, but only within measurable limits. The framework links internal representations to external recommendations, validated with random controls and mechanism boundaries. As LLMs become autonomous financial agents, this is a first step toward a behavioral layer for emerging know-your-agent (KYA) standards: knowing what an agent prefers, and how far that preference can be moved.

We propose a five-step diagnostic protocol for residual-trained neural HJB-PIDE solvers with control-dependent Lévy jumps, targeting a general failure mode of neural PDE methods: a learned solution can match headline scalar diagnostics while miscomputing an operator inside its training loss. The protocol pairs each neural solve with at least one from-scratch independent reference, decomposes the Hamiltonian into drift, diffusion, compensator, and nonlocal-integral components across a u-grid, and compares the value function and its low-order derivatives over a (t,x) grid before any argmax comparison. Applied to a standard CRRA-Merton-Variance-Gamma benchmark, it isolates a missing 1/2-mixture factor in the neural method's importance-proposal density that scaled the nonlocal integral by exactly half - a textbook signature of a constant proposal scale error, invisible to longer training, grid refinement, and truncation sweeps. With the bug corrected, four references - two finite-difference solvers with disjoint discretizations, the neural solver, and a semi-analytic scalar baseline obtained from CRRA homogeneity - agree on the optimal control to within ~2%. The constant-coefficient CRRA benchmark collapses by homogeneity to a scalar maximization, so the scalar baseline is the efficient method here; the contribution is the protocol, applicable in principle to non-homogeneous and higher-dimensional settings where neural HJB-PIDE solvers are genuinely needed. The episode is a concrete instance of a broader neural-PDE verification failure: pointwise agreement of a learned value or control can coexist with a systematically wrong nonlocal operator, so per-component and surface-level checks are needed before trusting the argmax policy.

Real-world asset tokenization is often presented as a mechanism for improving the liquidity of traditionally illiquid assets. However, on-chain representation and secondary-market liquidity are distinct outcomes. This paper examines whether tokenized real-world assets exhibit meaningful observed liquidity and identifies the token characteristics associated with higher market activity. Using token-level data from this http URL and supplemental contract-level observations from Etherscan, the study constructs an Ethereum-based monthly panel of non-stablecoin real-world assets across three prominent categories: U.S. Treasury-backed tokens, gold-backed commodity tokens, and private-credit-related tokens. Liquidity is measured using turnover, active addresses, and an active-month indicator. The empirical design combines descriptive statistics, non-parametric group tests, and exploratory panel regressions suited to short and sparse token histories. The results show substantial heterogeneity across asset categories. Gold-backed tokens exhibit broader holder bases and more persistent on-chain activity than many Treasury and private-credit-related products, while outstanding asset value alone does not reliably predict observed liquidity. The paper contributes to the literature by developing a clearer empirical measurement framework for real-world-asset liquidity and showing that tokenization and liquidity should be analyzed as distinct outcomes.

We provide a brief primer for the idea behind formalising hierarchical causality in the context of complex systems. Here actors are not simply agents. Actors instantiate causation classes. Agents implement local dynamics in given levels or organisation in a given system. Hierarchical causality then describes how actor-level roles constrain, select, and organise agent-level behaviour across levels. The system then necessarily requires three additional structures. First, causation classes to abstract a given form of causal influence that an actor instantiates. Second, aggregation operators to move across the levels. Third, discrete event-time maps are required because the system comprises events, and the relation between local event counts and any global clock must be specified. Our formulation here is purposefully simple and discrete.

This paper proposes a unified adaptive portfolio-management framework that combines factor-based view generation, Black-Litterman (BL) posterior estimation, EWMA covariance estimation, and mean-variance optimization. The key mechanism is a dynamic sliding window that adjusts the estimation horizon according to realized portfolio volatility, thereby updating factor estimates, BL posterior expected returns, and portfolio weights over time. In a ten-year empirical study of the top 100 market-capitalization constituents of the S&P 500 with turnover transaction costs, the proposed method outperforms dynamic mean-variance optimization without BL views and provides stronger downside risk control, while its relative performance remains benchmark-dependent.

Major government studies and policy reports project that substantial expansion of interregional transmission will be needed to integrate clean energy and ensure reliability in decarbonized power systems. Using the open-source Switch capacity expansion model with detailed representation of existing U.S. generation and transmission infrastructure, solar, wind, and storage resources, and hourly operations, we evaluate the role of interregional transmission across least-cost, carbon-priced, and zero-emissions scenarios for 2050. An optimal nationwide plan would more than triple interregional transmission capacity, yet this reduces the cost of a zero emissions system by only 7% relative to relying on existing interregional transmission, as storage, solar and wind siting, and nuclear generation serve as close substitutes. Regional cost and rent effects vary, with transmission generally favoring wind and hydrogen resources over solar and batteries. Sensitivity analysis shows diminishing returns: one-fifth of the benefits of full expansion can be achieved with one-twelfth of the added capacity, while cost reductions for batteries and hydrogen provide comparable or greater system savings than interregional transmission. Upgrading existing interregional corridors with advanced conductors roughly doubling capacity per link at half the cost of new builds reduces system costs by only 1.6%, suggesting that reconductoring benefits are modest and that realizing their full potential likely requires pairing with new connections on key corridors or complementary reductions in battery costs. These results suggest that while substantial transmission expansion is economically justified, a diverse set of flexibility resources can substitute for large-scale grid build out, and the relative value of transmission is highly contingent on technological and cost developments.

We introduce a framework for simulating macroeconomic expectations in survey experiments using LLM-based economic agents (LLM Agents). We construct LLM Agents equipped with several functional modules that retrieve personal characteristics, prior expectations, and dynamic external information. We validate our framework by recapitulating three representative survey designs covering various expectations across different types of respondents. Our results show that LLM Agents generate expectation distributions highly similar to human data and capture human-aligned qualitative patterns in open-ended responses. Evaluation reveals that priors are crucial for matching distributions, whereas personal and external information drive human-like thought processes. Our findings offer guidance for narrowing the belief gap between generative AI and humans at the aggregate level while delineating the boundaries of the framework.

This paper studies the pricing problem in which the underlying asset follows a non-Markovian stochastic volatility model. Classical partial differential equation methods face significant challenges in this context, as the option prices depend not only on the current state, but also on the entire historical path of the process. To overcome these difficulties, we reformulate the asset dynamics as a rough stochastic differential equation and then represent the rough paths via linear or non-linear combinations of time-extended Brownian motion signatures. This representation transforms a rough stochastic differential equation to a classical stochastic differential equation, allowing the application of standard analytical tools. We propose a deep signature approach for both linear and nonlinear representations and rigorously prove the convergence of the algorithm. Numerical examples demonstrate the effectiveness of our approach for both Markovian and non-Markovian volatility models, offering a theoretically grounded and computationally efficient framework for option pricing.

We quantify the short-term impact of Generative Artificial Intelligence (GenAI) on sales performance through a series of large-scale randomized field experiments involving millions of users and products at a leading cross-border online retail platform. Over 2023-2024, the platform integrated GenAI into seven consumer-facing business workflows spanning customer service, consumer-product matching, advertising, and seller services. We find that GenAI adoption increases sales in most workflows, with effects ranging from no detectable impact to $16.3\%$, depending on GenAI's marginal contribution relative to baseline firm practices. Across the four GenAI applications with positive sales effects, the implied annual incremental value is roughly $\$5-$an economically meaningful impact given the retailer's scale and the early stage of GenAI adoption. The gains operate primarily through higher conversion rates rather than larger cart values, consistent with GenAI improving the shopping experience by reducing search, information, communication, and personalization frictions. Importantly, these effects are not associated with worse post-purchase outcomes, as product return rates and customer ratings do not deteriorate. Finally, we document substantial demand-side heterogeneity, with larger gains for less experienced consumers. Our findings provide novel, large-scale causal evidence on how GenAI shapes sales productivity in online retail, highlighting both its immediate value and broader potential.

The paper is related to the identification of firm's features which serve as determinants for firm's total factor productivity through unsupervised learning techniques (principal component analysis, self organizing maps, clustering). This bottom-up approach can effectively manage the problem of the heterogeneity of the firms and provides new ways to look at firms' standard classifications. Using the large sample provided by the ORBIS database, the analyses covers the years before the outbreak of Covid-19 (2015-2019) and the immediate post-Covid period (year 2020). It has been shown that in both periods, the main determinants of productivity growth are related to profitability, credit/debts measures, cost and capital efficiency, and effort/efficiency of the R&D activity conducted by the firms. Finally, a linear relationship between determinants and productivity growth has been found.

In financial and actuarial research, distortion and Haezendonck-Goovaerts risk measures are attractive due to their strong properties. They have so far been treated separately. In this paper, following a suggestion by Goovaerts, Linders, Van Weert, and Tank, we introduce and study a new class of risk measure that encompasses the distortion and Haezendonck-Goovaerts risk measures, aptly called the distortion Haezendonck-Goovaerts risk measures. They will be defined on a larger space than the space of bounded risks. We provide situations where these new risk measures are coherent, and explore their risk theoretic properties.

We analyze a continuous-time preemption game with shared catastrophic externalities. When the cost of catastrophe is embedded in both players' payoffs, the risk term cancels out in the equilibrium indifference condition. This creates a "suicide region" where competitive pressures force rational agents to deploy despite negative risk-adjusted net present values. We apply this framework to the race for artificial general intelligence (AGI). We show that this suicide region widens as the cost of systemic ruin grows: higher catastrophic risk does not deter the race but instead enlarges the set of conditions under which rational actors deploy despite negative social value. We characterize the resulting welfare distortion against a social planner's benchmark and demonstrate how two complementary mechanisms - private liability and prize-sharing - can close the suicide region. Private liability raises the cost of unsafe deployment while prize-sharing reduces the strategic imperative to deploy first. "Warning shots" (sub-existential disasters) will fail to deter AGI acceleration, as the winner-takes-all nature of the race remains intact.

Risk forecasts in financial regulation and internal management are calculated through historical data. The unknown structural changes of financial data pose a substantial challenge in selecting an appropriate look-back window for risk modeling and forecasting. We develop a data-driven online learning method, called the bootstrap-based adaptive window selection (BAWS), that adaptively determines the window size in a sequential manner. A central component of BAWS is to compare the realized scores against a data-dependent threshold based on the bootstrap method. We provide an asymptotic justification for the bootstrap threshold, covering non-smooth scores such as the VaR check loss and the joint VaR--ES score, with an extension to stationary weakly dependent data via the moving block bootstrap. A single-break analysis further shows that BAWS rejects overlong windows crossing sufficiently large breaks. The proposed method is applicable to the forecasting of risk measures that are elicitable individually or jointly, such as the Value-at-Risk (VaR) and the pair of VaR and the corresponding Expected Shortfall. Through simulation studies and an empirical analysis, we demonstrate that BAWS often improves upon the standard rolling window approach and the recently developed method of stability-based adaptive window selection, especially when there are structural changes in the data-generating process.

Can contagion components be identified in aggregated default counts when default probabilities fluctuate with the economic environment? We study this question as an identifiability problem for coarse-grained default-count distributions. Three dependence mechanisms are compared: cumulative contagion in the Davis--Lo model, threshold-type contagion in the Torri model, and common-factor dependence in the Vasicek model. Under an i.i.d. specification, the Vasicek model gives the best overall fit, indicating that a smooth mixture induced by environmental fluctuations can reproduce much of the observed annual default clustering. We then introduce a hierarchical specification in which the baseline default probability varies across years. This extension separates cross-year environmental fluctuations from within-year contagion. Most of the variance of annual default counts is explained by fluctuations in default conditions. The remaining component, however, depends on the contagion mechanism. Threshold-type contagion is largely absorbed into environmental heterogeneity, whereas cumulative contagion leaves a small but persistent signature in both variance decomposition and tail behavior. These results clarify when contagion remains identifiable after aggregation and when it becomes indistinguishable from environmental fluctuations.

We study risk-neutral density extraction from short-dated option chains. As expiry approaches, option premia decline and bid--ask spreads can be large relative to prices, making mid quotes particularly uninformative. Stale or asynchronous quotes may also generate potential static arbitrages, rendering standard procedures infeasible or unstable. We develop a model-free pipeline that treats bid-ask quotes as the primitive market constraint. The pipeline consists of two steps. First, a procedure called ``Arbitrage Removal Iterative Executable Strategy'' (ARIES) filters executable static arbitrage at quoted bid and ask prices under market-depth constraints. Second, the ``Smooth Entropic Density EXtraction'' (SEDEx) then recovers the density through a criterion leveraging smoothness and entropy under bid-ask constraints. We test the pipeline on synthetic Heston panels and short-dated SPX option data, sampled from a few hours to one week before expiry. Computation is fast and returns robust densities across various market conditions, including scheduled macroeconomic announcements. As an empirical application, we use the recovered densities to construct short dated implied-volatility smiles.

Like physical products, new technologies are developed using globally sourced inputs. Yet while the supply chains behind physical goods are well understood, we know far less about the international supply chain of scientific knowledge that powers U.S. innovation, or how vulnerable it may be to disruption. Here, I uncover this supply chain by tracing multi-generational citation paths connecting NSF-funded research to downstream patents, and stress-test it by simulating barriers to scientific knowledge flows across the U.S. border. The U.S. knowledge supply chain extends globally, and frictions impeding the movement of ideas across the U.S. border reduce its connectivity, extend its length, and lower innovation productivity. These impacts extend to technology areas deemed critical to national priorities by U.S. Congress, including Semiconductors, Quantum Science, and AI.

This paper develops a valuation framework for guaranteed lifetime withdrawal benefit (GLWB) contracts with long-term care (LTC) features when the reference fund follows exponential Levy dynamics and the short rate follows the Hull-White model. The contract combines financial guarantees, longevity protection, health-contingent LTC payments, and surrender optionality, requiring the joint treatment of jump risk, stochastic discounting, and disability risk. The numerical method couples a recombining Hull-White trinomial tree with an implicit-explicit (IMEX) finite difference scheme. The framework incorporates a seven-state health model, annual fees, LTC payments, guaranteed withdrawals, and bang-bang policyholder actions, and is benchmarked against Monte Carlo simulation. Numerical results show that the hybrid tree-IMEX method delivers stable long-maturity prices consistent with simulation benchmarks. They also show that Levy equity dynamics and stochastic interest rates have a material impact on fair fees and surrender incentives, and affect the decomposition of contract value. The findings highlight the importance of modelling financial tail risk and interest-rate risk jointly when pricing long-term insurance guarantees with LTC-contingent benefits.

We study a multi-factor block model for variable clustering and connect it to regularized subspace clustering through a distributionally robust version of nodewise regression. To solve the latter problem, we derive a convex relaxation, provide a data-driven approach for selecting the size of the robust region, and develop an ADMM algorithm for efficient implementation. We validate our method in extensive numerical studies and demonstrate its superior performance.

We study a monopolistic insurance market with hidden information, where the agent's type $\theta$ is private information that is unobservable to the insurer, and it is drawn from a continuum of types. The hidden type affects both the loss distribution and the risk attitude of the agent. Within this framework, we show that a menu of contracts is incentive efficient if it maximizes social welfare function, subject to incentive compatibility and individual rationality constraints. This holds for general utility functionals. In the special case of Yaari Dual Utility, we provide two partial converse statements to this result, and we give a semi-explicit characterization of optimal solutions to the social welfare maximization problem. We do this under two different settings: (i) the first assumes that types are ordered in a way such that larger values of $\theta$ correspond to more risk-averse types who face stochastically larger losses; whereas (ii) the second assumes that larger values of $\theta$ correspond to less risk-averse types who face stochastically larger losses. In both settings, the structure of optimal menus of contracts depends on the level of the social welfare weight, and we examine several properties thereof.

Privacy-preserving cryptocurrency exchanges alter what the pricing mechanism observes about order flow. We derive the unique linear Kyle equilibrium when a committed Bayesian market maker observes order flow perturbed by independent Gaussian privacy noise. The price-impact coefficient and informed-trader strategy rescale by reciprocal factors of the privacy parameter (one down, one up), so their product is invariant. A welfare decomposition then identifies a closed-form per-period transfer from the protocol's LP pool to traders -- the "privacy subsidy", the break-even fee any privacy-aggregated exchange must charge. The result is the single-period closed-form privacy-noise analog of Loss-Versus-Rebalancing (Milionis et al. 2022). The primary application is shielded AMMs with explicit additive-noise injection (e.g., differential privacy); related designs (batched swaps, sealed-bid auctions, oracle-pegged crossings) require separate frameworks that we leave to future work.

Cardinality-constrained portfolio selection is routinely cast as a quadratic unconstrained binary optimization (QUBO) and submitted to a quantum processing unit (QPU) for direct annealing. We show that this standard penalty encoding is the binding constraint for direct-QPU execution on current D-Wave Pegasus and Zephyr hardware. Expanding the exact cardinality penalty contributes a dense rank-one term that makes the logical interaction graph complete regardless of the covariance, producing chain-break fractions from 83% at small universes up to 92% at the full forty-nine-industry Fama--French universe, and zero feasible raw samples at every tested scale. Topology-aware sparsification reduces chain breaks to near zero, but any sparsifier that removes off-diagonal entries also dilutes the cardinality constraint; an ablation reveals that this sparsify-and-project pipeline is dominated by the classical projector, not the QPU. We propose removing the penalty entirely: sample an objective-only QUBO built from expected returns and the risk-scaled covariance on hardware, and enforce cardinality classically through a deterministic feasibility projector. Across 4,468 saved embedding records on live Pegasus and Zephyr hardware, spanning equities up to forty-nine assets and football-betting instances up to forty-eight, this penalty-free pipeline reduces mean chain-break fractions from 71%--92% down to at most 0.04%, and post-processed regret is at most 0.03% relative to greedy classical references at every tested scale. We do not claim quantum advantage; the penalty encoding, not the sparse hardware topology, is the limiting factor for direct-QPU portfolio optimization at currently accessible scales.

We derive a closed-form bid-ask spread and welfare decomposition for the Glosten-Milgrom 1985 sequential-trading model when the market maker observes the trade direction perturbed by a binary flip channel of probability $\eta$ -- a natural information-theoretic model of privacy mechanisms acting on the direction signal. Under a committed Bayesian market-maker pricing rule, the equilibrium spread is $\mu(1-2\eta)\Delta$, where $\mu$ is the informed-trader fraction and $\Delta = v_H - v_L$ the value range. The welfare decomposition identifies a per-trade transfer $\mu\eta\Delta$ from the protocol's liquidity pool to traders -- the "privacy subsidy", mirroring the Gaussian-Kyle analog established in prior work. The result extends the privacy-subsidy concept from continuous Gaussian to discrete two-state microstructure, demonstrating robustness across both classical models. Primary application: MPC-based matching engines with $\varepsilon$-differentially-private direction disclosure, where the engine prices on a noisy direction signal.

We extend the closed-form privacy-subsidy result of Nakamura~(2026, arXiv:2605.15746) from the single-period Kyle model to continuous-time. A committed Bayesian automated market maker observes the aggregate order flow perturbed by an independent Brownian privacy channel of diffusion intensity $\sigma_\varepsilon$. Under the Markovian linear equilibrium, the price-impact coefficient is $\lambda = \sigma_v / \sqrt{\sigma_u^2 + \sigma_\varepsilon^2}$ -- constant in time -- and the cumulative expected transfer from the protocol's liquidity pool to traders over $[0,1]$ is $|\Pi_M| = \sigma_v \sigma_\varepsilon^2 / \sqrt{\sigma_u^2 + \sigma_\varepsilon^2}$. We then establish a structural correspondence between this cumulative privacy subsidy and Loss-Versus-Rebalancing (Milionis et al.~2022), identifying privacy-noise welfare as the order-flow observation analog of LVR's price observation gap. The result completes the continuous-time Kyle leg of the program of quantifying break-even fees for committed-AMM exchanges under privacy-aggregated information environments.