The COVID-19 pandemic exposed critical vulnerabilities in vaccine supply chains, highlighting the need for robust manufacturing for rapid pandemic response to support CEPI's 100 Days Mission. We develop a discrete-event simulation model to analyze supply chain disruptions and enables policymakers and vaccine manufacturers to quantify disruptions and assess mitigation strategies. Unlike prior studies examining components in isolation, our approach integrates production processes, quality assurance and control (QA/QC) activities, and raw material procurement to capture system-wide dynamics. A detailed mRNA case study analyzes disruption scenarios for a facility targeting 50 million doses: facility shutdowns, workforce reductions, raw material shortages, infrastructure failures, extended procurement lead times, and increased QA/QC capacity. Three main insights emerge. First, QA/QC personnel are the primary bottleneck, with utilization reaching 84.5% under normal conditions while machine utilization remains below 33%. Doubling QA/QC capacity increases annual output by 79.1%, offering greater returns than equipment investments. Second, raw material disruptions are highly detrimental, with extended lead times reducing three-year output by 19.6% and causing stockouts during 51.8% of production time. Third, the model shows differential resilience: acute disruptions (workforce shortages, shutdowns, power outages) allow recovery within 6 to 9 weeks, whereas chronic disruptions (supply delays) cause prolonged performance degradation.

Many economic models feature moment conditions that involve latent variables. When the latent variables are individual fixed effects in an auxiliary panel data regression, we construct orthogonal moments that eliminate first-order bias induced by estimating the fixed effects. Machine Learning methods and Empirical Bayes methods can be used to improve the estimate of the nuisance parameters in the orthogonal moments. We establish a central limit theorem based on the orthogonal moments without relying on exogeneity assumptions between panel data residuals and the cross-sectional moment functions. In a simulation study where the exogeneity assumption is violated, the estimator based on orthogonal moments has smaller bias compared with other estimators relying on that assumption. An empirical application on experimental site selection demonstrates how the method can be used for nonlinear moment conditions.

A major challenge in data-driven decision-making is accurate policy evaluation-i.e., guaranteeing that a learned decision-making policy achieves the promised benefits. A popular strategy is model-based policy evaluation, which estimates a model from data to infer counterfactual outcomes. This strategy is known to produce unwarrantedly optimistic estimates of the true benefit due to the winner's curse. We searched the recent literature on data-driven decision-making, identifying a sample of 55 papers published in the Management Science in the past decade; all but two relied on this flawed methodology. Several common justifications are provided: (1) the estimated models are accurate, stable, and well-calibrated, (2) the historical data uses random treatment assignment, (3) the model family is well-specified, and (4) the evaluation methodology uses sample splitting. Unfortunately, we show that no combination of these justifications avoids the winner's curse. First, we provide a theoretical analysis demonstrating that the winner's curse can cause large, spurious reported benefits even when all these justifications hold. Second, we perform a simulation study based on the recent and consequential data-driven refugee matching problem. We construct a synthetic refugee matching environment (calibrated to closely match the real setting) but designed so that no assignment policy can improve expected employment compared to random assignment. Model-based methods report large, stable gains of around 60% even when the true effect is zero; these gains are on par with improvements of 22-75% reported in the literature. Our results provide strong evidence against model-based evaluation.

We study whether a social planner can improve the efficiency of learning, measured by the expected total welfare loss, in a sequential decision-making environment. Agents arrive in order and each makes a binary action based on their private signal and the social information they observe. The planner can intervene by jointly designing the social information disclosed to agents and offering monetary transfers contingent on agents' actions. We show that, despite such flexibility, efficient learning cannot be restored with a finite budget: whenever learning is inefficient without intervention, no combination of information disclosure and transfers can achieve efficient learning while keeping total expected transfers finite.

Growing concerns about housing affordability have prompted the adoption of rent control policies and renewed debates over their effectiveness. This paper provides the first empirical evaluation of the 2024 rent control policy implemented in Catalonia under Spain's new national housing law. To identify the causal effect of the policy on the rental market, I use municipality-level administrative data and implement several difference-in-differences strategies and event study designs. The results point to a reduction in tenancy agreements and a less robust decrease in rental price growth. While the findings highlight important short-term consequences of rent control, they also underscore the need for caution due to data limitations and limited robustness in some estimates.

Around three quarters of Bitcoin transactions take place off-chain. Despite their significance, the vast majority of the empirical literature on cryptocurrencies focuses on on-chain transactions. This paper presents one of the first analysis of both on- and off-chain demand- and supply-side factors. Two hypotheses relating on-chain and off-chain demand and supply drivers to the Bitcoin price are tested in an ARDL model with daily data from 2019 to 2024. Our estimates document the differential contributions of on-chain and off-chain drivers on the Bitcoin price. Off-chain demand pressures have a significant impact on the Bitcoin price in the long-run. In the short-run, both demand and supply drivers significantly affect the Bitcoin price. Regarding transactions on the blockchain, only on-chain demand pressures are statistically significant - both in the long- and short-run. These findings confirm the dual nature of the Bitcoin price dynamics, where also market fundamentals affect the Bitcoin price in addition to speculative drivers. Bitcoin whale trading has less significant impact on price in the long-run, while is more pronounced contemporaneously and one-period lag.

This paper examines how gender and residential socioeconomic status shape hiring outcomes in the information technology sector using a field experiment from the city of Karachi, Pakistan. Employers in Pakistan can openly state preferences regarding gender, residential location, and other characteristics, but the majority in the information technology sector choose not to do so. This creates an opportunity to examine whether discrimination persists when such biases are not explicitly stated. An analysis of explicitly gender-targeted job ads shows that men are preferred over women across most occupations, even in traditionally pink-collar roles. Moreover, results from a resume audit experiment, submitting 2,032 applications to 508 full-time job openings, show that men receive more callbacks for job interviews than women, even in the absence of explicit gender preferences in job ads. The study also indicates a significant premium favoring candidates from high-income areas, who receive 45 percent more callbacks than applicants from low-income neighborhoods. This advantage remains robust even after controlling for commuting distance. Qualitative interviews with human resource officials suggest that employers associate productivity with both gender and neighborhood socioeconomic status. Residential address acts as a proxy for class background and signals education, skills, and perceived "fit" in professional settings. These perceptions may reinforce stereotypes, disadvantaging women and candidates from low-income backgrounds.

The mixed logit model is a flexible and widely used demand model in pricing and revenue management. However, existing work on mixed-logit pricing largely focuses on unconstrained settings, limiting its applicability in practice where prices are subject to business or regulatory constraints. We study the constrained pricing problem under multinomial and mixed logit demand models. For the multinomial logit model, corresponding to a single customer segment, we show that the constrained pricing problem admits a polynomial-time approximation scheme (PTAS) via a reformulation based on exponential cone programming, yielding an $\varepsilon$-optimal solution in polynomial time. For finite mixed logit models with $T$ customer segments, we reformulate the problem as a bilinear exponential cone program with $O(T)$ bilinear terms. This structure enables a Branch-and-Bound algorithm whose complexity is exponential only in $T$. Consequently, constrained pricing under finite mixtures of logit admits a PTAS when the number of customer segments is bounded. Numerical experiments demonstrate strong performance relative to state-of-the-art baselines.

In Bayesian single-item auctions, a monotone bidding strategy--one that prescribes a higher bid for a higher value type--can be equivalently represented as a partition of the quantile space into consecutive intervals corresponding to increasing bids. Kumar et al. (2024) prove that agile online gradient descent (OGD), when used to update a monotone bidding strategy through its quantile representation, is strategically robust in repeated first-price auctions: when all bidders employ agile OGD in this way, the auctioneer's average revenue per round is at most the revenue of Myerson's optimal auction, regardless of how she adjusts the reserve price over time. In this work, we show that this strategic robustness guarantee is not unique to agile OGD or to the first-price auction: any no-regret learning algorithm, when fed gradient feedback with respect to the quantile representation, is strategically robust, even if the auction format changes every round, provided the format satisfies allocation monotonicity and voluntary participation. In particular, the multiplicative weights update (MWU) algorithm simultaneously achieves the optimal regret guarantee and a strong strategic robustness guarantee in this auction setting. At a technical level, our results are established via a simple relation that bridges Myerson's auction theory and standard no-regret learning theory.

This article identifies the factors that drove house prices in 13 advanced countries over the past 35 years. It does so based on Breiman s (2001) random forest model. Shapley values indicate that annual house price growth across countries is explained first and foremost by price momentum, initial valuations (proxied by price to rent ratios) and household credit growth. Partial effects of explanatory variables are also elicited and suggest important non-linearities, for instance as to what concerns the effects of CPI inflation on house price growth. The out-of-sample forecast test reveals that the random forest model delivers 44% lower house price variation root square mean errors (RMSEs) and 45% lower mean absolute errors (MAEs) when compared to an OLS model that uses the same set of 10 pre-determined explanatory variables. Notably, the same model works well for all countries, as the random forest attributes minimal values to country fixed effects.

This paper analyzes a society composed of individuals who have diverse sets of beliefs (or models) and diverse tastes (or utility functions). It characterizes the model selection process of a social planner who wishes to aggregate individuals' beliefs and tastes but is concerned that their beliefs are misspecified (or incorrect). A novel impossibility result emerges under several desiderata: a utilitarian social planner who prioritizes robustness to misspecification never aggregates individuals' beliefs but instead behaves as a dictator by adopting one individual's belief as the social belief. This tension between robustness and aggregation exists because aggregation yields policy-contingent beliefs, which are very sensitive to policy outcomes. The impossibility can be resolved, but it would require assuming individuals have heterogeneous tastes and some common beliefs. Applications in treatment choice and dynamic macroeconomics are explored.

We axiomatically define a cardinal social inefficiency function, which, given a set of alternatives and individuals' vNM preferences over the alternatives, assigns a unique number -- the social inefficiency -- to each alternative. These numbers -- and not only their order -- are uniquely defined by our axioms despite no exogenously given interpersonal comparison, outside option, or disagreement point. We interpret these numbers as per-capita losses in endogenously normalized utility. We apply our social inefficiency function to a setting in which interpersonal comparison is notoriously hard to justify -- object allocation without money -- leveraging techniques from computer science to prove an approximate-efficiency result for the Random Serial Dictatorship mechanism.

We introduce a new smooth transition vector autoregressive model with a Gaussian conditional distribution and transition weights that, for a $p$th order model, depend on the full distribution of the preceding $p$ observations. Specifically, the transition weight of each regime increases in its relative weighted likelihood. This data-driven approach facilitates capturing complex switching dynamics, enhancing the identification of gradual regime shifts. In an empirical application to the macroeconomic effects of a severe weather shock, we find that in monthly U.S. data from 1961:1 to 2022:3, the shock has stronger impact in the regime prevailing in the early part of the sample and in certain crisis periods than in the regime dominating the latter part of the sample. This suggests overall adaptation of the U.S. economy to severe weather over time.

In an era of data abundance, statistical evidence is increasingly critical for business and policy decisions. Yet, organizations lack empirical tools to assess the value of evidence-based decision making (EBDM), optimize statistical precision, and balance the costs of evidence-gathering strategies against their benefits. To tackle these challenges, this article introduces an empirical framework to estimate the value of EBDM and evaluate the return on investment in statistical precision and project ideation. The framework leverages parametric and nonparametric empirical Bayes methods to account for parameter heterogeneity and measure how statistical precision changes the value of evidence. The value extracted from statistical evidence depends critically on how organizations translate evidence into policy decisions. Commonly used decision rules based on statistical significance can leave substantial value unrealized and, in some cases, generate negative expected value.

This paper adapts ideas from social identity theory to set out a new framework for modelling conspicuous consumption. Agents derive utility from their consumption of a status good and from belonging to an identity group with high status good consumption. Importantly, these two sources of utility are substitutes. Agents also feel pressure to conform with their neighbours in a network. This framework can rationalise a set of seemingly conflicting stylised facts about conspicuous consumption that are currently explained by different families of models. In addition, our model delivers new testable predictions regarding the effect of network structure and income inequality on conspicuous consumption.

A new mixture vector autoregressive model based on Gaussian and Student's $t$ distributions is introduced. As its mixture components, our model incorporates conditionally homoskedastic linear Gaussian vector autoregressions and conditionally heteroskedastic linear Student's $t$ vector autoregressions. For a $p$th order model, the mixing weights depend on the full distribution of the preceding $p$ observations, which leads to attractive practical and theoretical properties such as ergodicity and full knowledge of the stationary distribution of $p+1$ consecutive observations. A structural version of the model with statistically identified shocks is also proposed. The empirical application studies the effects of the Euro area monetary policy shock. We fit a two-regime model to the data and find the effects, particularly on inflation, stronger in the regime that mainly prevails before the Financial crisis than in the regime that mainly dominates after it. The introduced methods are implemented in the accompanying R package gmvarkit.

A structural Gaussian mixture vector autoregressive model is introduced. The shocks are identified by combining simultaneous diagonalization of the reduced form error covariance matrices with constraints on the time-varying impact matrix. This leads to flexible identification conditions, and some of the constraints are also testable. The empirical application studies asymmetries in the effects of the U.S. monetary policy shock and finds strong asymmetries with respect to the sign and size of the shock and to the initial state of the economy. The accompanying CRAN distributed R package gmvarkit provides a comprehensive set of tools for numerical analysis.

We introduce a new mixture autoregressive model which combines Gaussian and Student's $t$ mixture components. The model has very attractive properties analogous to the Gaussian and Student's $t$ mixture autoregressive models, but it is more flexible as it enables to model series which consist of both conditionally homoscedastic Gaussian regimes and conditionally heteroscedastic Student's $t$ regimes. The usefulness of our model is demonstrated in an empirical application to the monthly U.S. interest rate spread between the 3-month Treasury bill rate and the effective federal funds rate.

In this paper we investigate panel regression models with interactive fixed effects. We propose two new estimation methods that are based on minimizing convex objective functions. The first method minimizes the sum of squared residuals with a nuclear (trace) norm regularization. The second method minimizes the nuclear norm of the residuals. We establish the consistency of the two resulting estimators. Those estimators have a very important computational advantage compared to the existing least squares (LS) estimator, in that they are defined as minimizers of a convex objective function. In addition, the nuclear norm penalization helps to resolve a potential identification problem for interactive fixed effect models, in particular when the regressors are low-rank and the number of the factors is unknown. We also show how to construct estimators that are asymptotically equivalent to the least squares (LS) estimator in Bai (2009) and Moon and Weidner (2017) by using our nuclear norm regularized or minimized estimators as initial values for a finite number of LS minimizing iteration steps. This iteration avoids any non-convex minimization, while the original LS estimation problem is generally non-convex, and can have multiple local minima.

Once failure is irreversible, continuation payoffs cannot be meaningfully aggregated across strategies that differ in their survival properties. Standard scalar evaluation sidesteps this by arbitrarily completing payoffs beyond termination, but such completions are extrinsic to the game form. This paper introduces continuation-performance decomposition (CPD), proving that any evaluation satisfying natural regularity conditions, such as failure-completion invariance, survival locality, and local expected-utility coherence -- must separate continuation from performance lexicographically. Continuation priority thus emerges as a consequence of well-posed evaluation, not as a behavioral assumption. We establish equivalence between CPD and the limit of games with diverging failure penalties, show that viability is a game-form invariant independent of payoffs, and apply the framework to bank runs: preemptive withdrawals reflect rational viability vetoes rather than coordination failure when continuation is distributively asymmetric. CPD resolves a representational problem, not a preference problem.

We develop a general framework for incorporating distributional preferences in market design. We identify the structural properties of these preferences that guarantee the path independence of choice rules. In decentralized settings, a greedy rule uniquely maximizes these preferences; in centralized markets, the associated deferred-acceptance mechanism uniquely implements them. This framework subsumes canonical models, such as reserves and matroids, while accommodating complex objectives involving intersectional identities that lie beyond the scope of existing approaches. Our analysis provides unified axiomatic foundations and comparative statics for a broad class of distributional policies.

Communication scene recognition has been widely applied in practice, but using deep learning to address this problem faces challenges such as insufficient data and imbalanced data distribution. To address this, we designed a weighted loss function structure, named FilterLoss, which assigns different loss function weights to different sample points. This allows the deep learning model to focus primarily on high-value samples while appropriately accounting for noisy, boundary-level data points. Additionally, we developed a matching weight filtering algorithm that evaluates the quality of sample points in the input dataset and assigns different weight values to samples based on their quality. By applying this method, when using transfer learning on a highly imbalanced new dataset, the accuracy of the transferred model was restored to 92.34% of the original model's performance. Our experiments also revealed that using this loss function structure allowed the model to maintain good stability despite insufficient and imbalanced data.

Orthogonal delay-Doppler division multiplexing (ODDM) is a promising modulation technique for reliable communications in high-mobility scenarios. However, the existing channel estimation frameworks for ODDM systems cannot achieve both high accuracy and low complexity simultaneously, due to the inherent coupling of delay and Doppler parameters. To address this problem, a two-dimensional (2D) hierarchical sparse Bayesian learning (HSBL) based channel estimation framework is proposed in this paper. Specifically, we address the inherent coupling between delay and Doppler dimensions in ODDM by developing a partially-decoupled 2D sparse signal recovery (SSR) formulation on a virtual sampling grid defined in the delay-Doppler (DD) domain. With the help of the partially-decoupled formulation, the proposed 2D HSBL framework first performs low-complexity coarse on-grid 2D sparse Bayesian learning (SBL) estimation to identify potential channel paths. Then, high-resolution fine grids are constructed around these regions, where an off-grid 2D SBL estimation is applied to achieve accurate channel estimation. Simulation results demonstrate that the proposed framework achieves performance superior to conventional off-grid 2D SBL with significantly reduced computational complexity.

Prophet inequalities bound the expected reward that can be obtained in a stopping problem by the optimal reward of its corresponding off-line version. We propose a systematic technique for deriving prophet inequalities for stopping problems associated with selecting a point in a polyhedron. It utilizes a reduced-form linear programming representation of the stopping problem. We illustrate the technique to derive a number of known results as well as some new ones. For instance, we prove a $\frac{1}{2}$-prophet inequality when the underlying polyhedron is an on-line polymatroid; one whose underlying submodular function depends upon the realized rewards. We also demonstrate a composition by the Minkowski sum property. If an $r-$ prophet inequality holds for polyhedra $P^1$ and $P^2$, it also holds for their Minkowski sum.

We study inference in models where a transformation of parameters exhibits first-order degeneracy -- that is, its gradient is zero or close to zero, making the standard delta method invalid. A leading example is causal mediation analysis, where the indirect effect is a product of coefficients and the gradient degenerates near the origin. In these local regions of degeneracy the limiting behaviors of plug-in estimators depend on nuisance parameters that are not consistently estimable. We show that this failure is intrinsic -- around points of degeneracy, both regular and quantile-unbiased estimation are impossible. Despite these restrictions, we develop minimum-distance methods that deliver uniformly valid confidence intervals. We establish sufficient conditions under which standard chi-square critical values remain valid, and propose a simple bootstrap procedure when they are not. We demonstrate favorable power in simulations and in an empirical application linking teacher gender attitudes to student outcomes.

Bank failures can stem from runs on otherwise solvent banks or from losses that render banks insolvent, regardless of withdrawals. Disentangling the relative importance of liquidity and solvency in explaining bank failures is central to understanding financial crises and designing effective financial stability policies. This paper reviews evidence on the causes of bank failures. Bank failures -- both with and without runs -- are almost always related to poor fundamentals. Low recovery rates in failure suggest that most failed banks that experienced runs were likely fundamentally insolvent. Examiners' postmortem assessments also emphasize the primacy of poor asset quality and solvency problems. Before deposit insurance, runs commonly triggered the failure of insolvent banks. However, runs rarely caused the failure of strong banks, as such runs were typically resolved through other mechanisms, including interbank cooperation, equity injections, public signals of strength, or suspension of convertibility. We discuss the policy implications of these findings and outline directions for future research.

Do leading LLM developers possess a proprietary ``secret sauce'', or is LLM performance driven by scaling up compute? Using training and benchmark data for 809 models released between 2022 and 2025, we estimate scaling-law regressions with release-date and developer fixed effects. We find clear evidence of developer-specific efficiency advantages, but their importance depends on where models lie in the performance distribution. At the frontier, 80-90% of performance differences are explained by higher training compute, implying that scale--not proprietary technology--drives frontier advances. Away from the frontier, however, proprietary techniques and shared algorithmic progress substantially reduce the compute required to reach fixed capability thresholds. Some companies can systematically produce smaller models more efficiently. Strikingly, we also find substantial variation of model efficiency within companies; a firm can train two models with more than 40x compute efficiency difference. We also discuss the implications for AI leadership and capability diffusion.

Building on the generalized hedonic-linear model of Pellegrino (2025), this paper studies optimal product differentiation when a representative consumer has preferences over product characteristics. Under multiproduct monopoly, the monopolist's choice of product characteristics is always aligned with the social planner's optimum, despite underproduction. By contrast, under oligopoly, multiple equilibria can arise that differ qualitatively in their patterns of characteristics design. We show that, while oligopoly equilibria exhibiting product differentiation yield higher welfare than those with product concentration, the degree of product differentiation under oligopoly remains below the socially optimal level. As a result, social welfare under oligopoly is typically lower than under monopoly, highlighting a key advantage of coordination in characteristics design. We extend the analysis to settings with overlapping ownership structures and show that common ownership can improve welfare by inducing firms to soften competition through increased product differentiation rather than output reduction.

Clinical trials shape medical evidence and determine who gains access to experimental therapies. Whether participation in these trials reflects the global burden of disease remains unclear. Here we analyze participation inequality across more than 62,000 randomized controlled trials spanning 16 major disease categories from 2000 to 2024. Linking 36.8 million trial participants to country-level disease burden, we show that global inequality in clinical trials participation is overwhelmingly shaped by country rather than disease burden. Country-level factors explain over 90% of variation in participation, whereas disease-specific effects contribute only marginally. Removing entire disease categories-including those traditionally considered underfunded-has little effect on overall inequality. Instead, participation is highly concentrated geographically, with a small group of countries enrolling a disproportionate share of participants across nearly all diseases. These patterns have persisted despite decades of disease-targeted funding and increasing alignment between research attention and disease burden within diseases. Our findings indicate that disease-vertical strategies alone cannot correct participation inequality. Reducing global inequities in clinical research requires horizontal investments in research capacity, health infrastructure, and governance that operate across disease domains.

Instrumental variable methods are fundamental to causal inference when treatment assignment is confounded by unobserved variables. In this article, we develop a general nonparametric causal framework for identification and learning with multi-categorical or continuous instrumental variables. Specifically, the mean potential outcomes and the average treatment effect can be identified via a regular weighting function derived from the proposed framework. Leveraging semiparametric theory, we derive efficient influence functions and construct two consistent, asymptotically normal estimators via debiased machine learning. The first estimator uses a prespecified weighting function, while the second estimator selects the optimal weighting function adaptively. Extensions to longitudinal data, dynamic treatment regimes, and multiplicative instrumental variables are further developed. We demonstrate the proposed method by employing simulation studies and analyzing real data from the Job Training Partnership Act program.