Federated learning (FL) is a machine learning technique that distributes model training to multiple clients while allowing clients to keep their data local. Although the technique allows one to break free from data silos keeping data local, to coordinate such distributed training, it requires an orchestrator, usually a central server. Consequently, organisational issues of governance might arise and hinder its adoption in both competitive and collaborative markets for data. In particular, the question of how to govern FL applications is recurring for practitioners. This research commentary addresses this important issue by inductively proposing a layered decision framework to derive organisational archetypes for FL’s governance. The inductive approach is based on an expert workshop and post-workshop interviews with specialists and practitioners, as well as the consideration of real-world applications. Our proposed framework assumes decision-making occurs within a black box that contains three formal layers: data market, infrastructure, and ownership. Our framework allows us to map organisational archetypes ex-ante. We identify two key archetypes: consortia for collaborative markets and in-house deployment for competitive settings. We conclude by providing managerial implications and proposing research directions that are especially relevant to interdisciplinary and cross-sectional disciplines, including organisational and administrative science, information systems research, and engineering.

Contact tracing is an effective public health policy to put the fast-spreading epidemic under control. The government tracks the contacts of confirmed SARS-CoV-2 cases, recommends testing, encourages self-quarantine, and monitors symptoms of contacts. In developing and less-developed countries with limited resources for widespread SARS-CoV-2 testing, it remains essential to identify and quarantine positive contacts to control outbreaks. Therefore, analysing recall and precision when implementing testing policies for these contacts is necessary. We analysed a contact tracing dataset from a cohort of 827 index patients infected with SARS-CoV-2 and their 14814 close contacts from Jan 2020 to July 2020 in a province in eastern China. We constructed a network from the data and used a Graph Convolutional Network to predict each contact’s infection status. To the best of our knowledge, this is the first method to use population-based contact tracing data for predicting the infection status using graph neural networks. Despite limited information, our model achieves competitive Area Under the Receiver Operating Characteristic Curve (ROC AUC) compared to hospital-onset scenarios. Based on the risk scores, we propose several contact testing policy adaptations that balance resource efficiency and effective pandemic control.

This study explores the potential of applying machine learning (ML) methods to identify and predict areas at risk of food insufficiency using a parsimonious set of publicly available data sources. We combine household survey data that captures monthly reported food insufficiency with remotely sensed measures of factors influencing crop production and maize price observations at the census enumeration area (EA) in Malawi. We consider three machine-learning models of different levels of complexity suitable for tabular data (TabNet, random forests, and LASSO) and classical logistic regression and examine their performance against the historical occurrence of food insufficiency. We find that the models achieve similar accuracy levels with differential performance in terms of precision and recall. The Shapley additive explanation decomposition applied to the models reveals that price information is the leading contributor to model fits. A possible explanation for the accuracy of simple predictors is the high spatiotemporal path dependency in our dataset, as the same areas of the country are repeatedly affected by food crises. Recurrent events suggest that immediate and longer-term responses to food crises, rather than predicting them, may be the bigger challenge, particularly in low-resource settings. Nonetheless, ML methods could be useful in filling important data gaps in food crises prediction, if followed by measures to strengthen food systems affected by climate change. Hence, we discuss the tradeoffs in training these models and their use by policymakers and practitioners.

In this article, we focus on the systemic expected shortfall and marginal expected shortfall in a multivariate continuous-time risk model with a general càdlàg process. Additionally, we conduct our study under a mild moment condition that is easily satisfied when the general càdlàg process is determined by some important investment return processes. In the presence of heavy tails, we derive asymptotic formulas for the systemic expected shortfall and marginal expected shortfall under the framework that includes wide dependence structures among losses, covering pairwise strong quasi-asymptotic independence and multivariate regular variation. Our results quantify how the general càdlàg process, heavy-tailed property of losses, and dependence structures influence the systemic expected shortfall and marginal expected shortfall. To discuss the interplay of dependence structures and heavy-tailedness, we apply an explicit order 3.0 weak scheme to estimate the expectations related to the general càdlàg process. This enables us to validate the moment condition from a numerical perspective and perform numerical studies. Our numerical studies reveal that the asymptotic dependence structure has a significant impact on the systemic expected shortfall and marginal expected shortfall, but heavy-tailedness has a more pronounced effect than the asymptotic dependence structure.

The basic principle of any version of insurance is the paradigm that exchanging risk by sharing it in a pool is beneficial for the participants. In case of independent risks with a finite mean, this is the case for risk-averse decision-makers. The situation may be very different in case of infinite mean models. In that case it is known that risk sharing may have a negative effect, which is sometimes called the nondiversification trap. This phenomenon is well known for infinite mean stable distributions. In a series of recent papers, similar results for infinite mean Pareto and Fréchet distributions have been obtained. We further investigate this property by showing that many of these results can be obtained as special cases of a simple result demonstrating that this holds for any distribution that is more skewed than a Cauchy distribution. We also relate this to the situation of deadly catastrophic risks, where we assume a positive probability for an infinite value. That case gives a very simple intuition why this phenomenon can occur for such catastrophic risks. We also mention several open problems and conjectures in this context.

For time series with high temporal correlation, the empirical process converges rather slowly to its limiting distribution. Many statistics in change-point analysis, goodness-of-fit testing, and uncertainty quantification admit a representation as functionals of the empirical process and therefore inherit its slow convergence. As a result, inference based on the asymptotic distribution of those quantities is significantly affected by relatively small sample sizes. We assess the quality of higher-order approximations (HOAs) of the empirical process by deriving the asymptotic distribution of the corresponding error terms. Based on the limiting distribution of the higher-order terms, we propose a novel approach to calculate confidence intervals for statistical quantities such as the median. In a simulation study, we compare coverage rates and lengths of these confidence intervals with those based on the asymptotic distribution of the empirical process and highlight some benefits of HOAs of the empirical process.

Data governance has emerged as a pivotal area of study over the past decade, yet despite its growing importance, a comprehensive analysis of the academic literature on this subject remains notably absent. This paper addresses this gap by presenting a systematic review of all academic publications on data governance from 2007 to 2024. By synthesizing insights from more than 3500 documents authored by more than 9000 researchers across various sources, this study offers a broad yet detailed perspective on the evolution of data governance research.

We consider the problem of sequential matching in a stochastic block model with several classes of nodes and generic compatibility constraints. When the probabilities of connections do not scale with the size of the graph, we show that under the Ncond condition, a simple max-weight type policy allows us to attain an asymptotically perfect matching while no sequential algorithm attains perfect matching otherwise. The proof relies on a specific Markovian representation of the dynamics associated with Lyapunov techniques.

Detecting structural changes in economic relationships has been a longstanding challenge in econometrics. Most of the literature on structural breaks has considered abrupt structural breaks. Existing tests for detecting smooth structural change typically rely on kernel estimation. In this article, we introduce a novel tuning-parameter-free test that minimizes a criterion function over all possible nondecreasing or nonincreasing structural change functions. This test is pivotal (after appropriate scaling) in the scalar case and remains computationally simple even in multivariate settings. Compared to existing nonparametric tests, our method offers superior power against local monotonic structural changes and does not involve the choice of a bandwidth parameter. A simulation study and two empirical examples highlight the merits of the proposed test relative to some popular tests for structural changes in the literature.

Antimicrobial resistance (AMR) in intensive care units (ICUs) is a critical issue, which has been exacerbated by the coronavirus disease 2019 (COVID-19) pandemic. This study investigated AMR prevalence and its associated factors among ICU patients in two Vietnamese hospitals from January 2020 to June 2022. Electronic medical records of 1,296 patients with 2,432 non-duplicate bacterial isolates were collected in Phu Tho Hospital (Northern, rural, non-COVID-19 treatment) and 175 Hospital (Southern, urban, COVID-19 treatment centre). Antibiotic susceptibility testing was conducted using VITEK2, BD Phoenix 100, and disk diffusion methods. Logistic regression with 1,000 bootstrap resampling and cross-validation was used to examine factors linked to AMR. Results revealed Acinetobacter spp. (27.5%) as leading strains in Phu Tho Hospital, while Klebsiella spp. (28.0%) predominated in 175 Hospital, except during 2021when Acinetobacter spp. reached the peak. Alarmingly, Acinetobacter spp., Klebsiella spp., and Pseudomonas aeruginosa demonstrated the highest AMR rates and multidrug resistance rates (83.8%–95.8%) in both hospitals. Resistance to cephalosporins, carbapenems, and fluoroquinolones ranged from 75% to 100%. Significant associated factors included age, sex, location, initial admission diagnosis, and bacterial isolation month. This study highlights the urgent need for controlling AMR in ICUs during the pandemic.

Mosquito-borne California serogroup orthobunyaviruses Inkoo (INKV) and Chatanga (CHATV) are known to be endemic in Finland with a high seroprevalence. We developed a novel multiplexed reverse transcription quantitative polymerase chain reaction method for discriminating between the INKV and CHATV. This assay was used along with traditional serological tests to study a set of summertime patients during the years 2021, 2023, and 2024 to assess the epidemiology and prevalence of acute INKV and CHATV infections in Finland. Altogether, 1470 samples were screened, and there were 16 patients who had an acute infection based on serological findings and/or nucleic acid test. The orthobunyavirus-IgG seroprevalences were 18% (2021), 20% (2023), and 30% (2024), being lower than that in studies from 20 years ago. Neutralization tests were carried out, and all but one acute case had more than four-fold higher titre to INVK vs. CHATV, indicating specificity to INKV infection. The results suggest that epidemiology has changed from previous studies, and INKV should be considered a causative agent of summertime infections in Finland. The symptom diversity in mild disease outcomes should be studied to guide orthobunyavirus recognition by clinicians. The use of molecular assay discriminating INKV and CHATV aids in understanding disease associations.

In recent decades, researchers have analyzed professional military education (PME) organizations to understand the characteristics and transformation of the core of military culture, the officer corps. Several historical studies have demonstrated the potential of this approach, but they were limited by both theoretical and methodological hurdles. This paper presents a new historical-institutionalist framework for analyzing officership and PME, integrating computational social science methods for large-scale data collection and analysis to overcome limited access to military environments and the intensive manual labor required for data collection and analysis. Furthermore, in an era where direct demographic data are increasingly being removed from the public domain, our indirect estimation methods provide one of the few viable alternatives for tracking institutional change. This approach will be demonstrated using web-scraping and a quantitative text analysis of the entire repository of theses from an elite American military school.

We revisit the recently introduced concept of return risk measures (RRMs) and extend it by incorporating risk management via multiple so-called eligible assets. The resulting new class of risk measures, termed multi-asset return risk measures (MARRMs), introduces a novel economic model for multiplicative risk sharing. We point out the connection between MARRMs and the well-known concept of multi-asset risk measures (MARMs). Then, we conduct a case study, based on an insurance dataset, in which we use typical continuous-time financial markets and different notions of acceptability of losses to compare RRMs, MARMs, and MARRMs and draw conclusions about the cost of risk mitigation. Moreover, we analyze theoretical properties of MARRMs. In particular, we prove that a positively homogeneous MARRM is quasi-convex if and only if it is convex, and we provide conditions to avoid inconsistent risk evaluations. Finally, the representation of MARRMs via MARMs is used to obtain various dual representations.

Tuberculosis (TB) remains a significant public health concern in China. Using data from the Global Burden of Disease (GBD) study 2021, we analyzed trends in age-standardized incidence rate (ASIR), prevalence rate (ASPR), mortality rate (ASMR), and disability-adjusted life years (DALYs) for TB from 1990 to 2021. Over this period, HIV-negative TB showed a marked decline in ASIR (AAPC = −2.34%, 95% CI: −2.39, −2.28) and ASMR (AAPC = −0.56%, 95% CI: −0.62, −0.59). Specifically, drug-susceptible TB (DS-TB) showed reductions in both ASIR and ASMR, while multidrug-resistant TB (MDR-TB) showed slight decreases. Conversely, extensively drug-resistant TB (XDR-TB) exhibited upward trends in both ASIR and ASMR. TB co-infected with HIV (HIV-DS-TB, HIV-MDR-TB, HIV-XDR-TB) showed increasing trends in recent years. The analysis also found an inverse correlation between ASIRs and ASMRs for HIV-negative TB and the Socio-Demographic Index (SDI). Projections from 2022 to 2035 suggest continued increases in ASIR and ASMR for XDR-TB, HIV-DS-TB, HIV-MDR-TB, and HIV-XDR-TB. The rising burden of XDR-TB and HIV-TB co-infections presents ongoing challenges for TB control in China. Targeted prevention and control strategies are urgently needed to mitigate this burden and further reduce TB-related morbidity and mortality.

This paper provides practical guidance to UK-based financial institutions (UKFIs) that are subject to the “operational resilience” guideline requirements of the Bank of England (BoE), Prudential Regulatory Authority and Financial Conduct Authority, issued in 2021, and fully effective for 31 March 2025. It contains practical suggestions and recommendations to assist UKFIs in implementing the guidelines. The scope of the paper covers issues related to (a) overviewing the latest equivalent operational resilience guidance in other countries and internationally (b) identifying key issues related to risk culture, risk appetite, information technology, tolerance setting, risk modelling, scenario planning and customer oriented operational resilience (c) identifying a framework for operational resilience based on a thorough understanding of these parameters and (d) designing and implementing an operational resilience maturity dashboard based on a sample of large UKIFs. The study also contains recommendations for further action, including enhanced controls and operational risk management frameworks. It concludes by identifying imperative policy actions to ensure that the implementation of the guidelines is more effective.

In this paper we adopt the probabilistic mean value theorem in order to study differences of the variances of transformed and stochastically ordered random variables, based on a suitable extension of the equilibrium operator. We also develop a rigorous approach aimed at expressing the variance of transformed random variables. This is based on a joint distribution which, in turn, involves the variance of the original random variable, as well as its mean residual lifetime and mean inactivity time. Then we provide applications to the additive hazards model and to some well-known random variables of interest in actuarial science. These deal with a new notion, called the ‘centred mean residual lifetime’, and a suitably related stochastic order. Finally, we also address the analysis of the differences of the variances of transformed discrete random variables thanks to the use of a discrete version of the equilibrium operator.

The gambler’s ruin problem for correlated random walks (CRWs), both with and without delays, is addressed using the optional stopping theorem for martingales. We derive closed-form expressions for the ruin probabilities and the expected game duration for CRWs with increments $\{1,-1\}$ and for symmetric CRWs with increments $\{1,0,-1\}$ (CRWs with delays). Additionally, a martingale technique is developed for general CRWs with delays. The gambler’s ruin probability for a game involving bets on two arbitrary patterns is also examined.

Cryptosporidium parvum is a well-established cause of gastrointestinal illness in both humans and animals and often causes outbreaks at animal contact events, despite the availability of a code of practice that provides guidance on the safe management of these events. We describe a large C. parvum outbreak following a lamb-feeding event at a commercial farm in Wales in 2024, alongside findings from a cohort study to identify high-risk exposures. Sixty-seven cases were identified, 57 were laboratory-confirmed C. parvum, with similar genotypes. Environmental investigations found a lack of adherence to established guidance. The cohort study identified 168 individuals with cryptosporidiosis-like illness from 540 exposure questionnaires (distributed via email to 790 lead bookers). Cases were more likely to have had closer contact with lambs (odds ratio (OR) kissed lambs = 2.4, 95% confidence interval (95% CI): 1.2–4.8). A multivariable analysis found cases were more likely to be under 10 years (adjusted OR (aOR) = 4.5, 95% CI: 2.0–10.0) and have had visible faeces on their person (aOR = 3.6, 95% CI: 2.1–6.2). We provide evidence that close contact at lamb-feeding events presents an increased likelihood of illness, suggesting that farms should limit animal contact at these events and that revisions to established codes of practice may be necessary. Enhancing risk awareness among farmers and visitors is needed, particularly regarding children.

In this paper we study the optimal multiple stopping problem with weak regularity for the reward, where the reward is given by a set of random variables indexed by stopping times. When the reward family is upper semicontinuous in expectation along stopping times, we construct the optimal multiple stopping strategy using the auxiliary optimal single stopping problems. We also obtain the corresponding results when the reward is given by a progressively measurable process.