Political scientists are increasingly interested in analyzing visual content at scale. However, the existing computational toolbox is still in need of methods and models attuned to the specific challenges and goals of social and political inquiry. In this article, we introduce a visual structural topic model (vSTM) that combines continuous image embeddings with a structural topic model. This has important advantages compared to existing approaches. First, continuous embeddings allow the model to capture the semantic complexity of images, which is relevant to the study of politics. A consequence of using continuous embeddings is that we can directly use off-the-shelf pretrained deep learning models, offering state-of-the-art performance in image analysis tasks while saving time, cost and resources. Second, the structural topic model provides the ability to analyze how topics and covariates are related, while maintaining a nuanced representation of images as a mixture of multiple topics. In our experiments, we show that the vSTM is able to identify topics that are interpretable, coherent and relevant to the study of online political communication.

Broadband Very Long Baseline Interferometry (VLBI) observations spanning 3–14 GHz pose calibration challenges that traditional methods were not designed to handle. Frequency-dependent instrumental responses, radio frequency interference (RFI), and source variability all complicate the estimation of station sensitivity and source flux density. We present a Bayesian hierarchical random-walk model that jointly tracks source flux density and baseline System Equivalent Flux Density (SEFD) from scan-level signal-to-noise ratio (SNR) observations. The model captures temporal changes via random walks, frequency dependence through piecewise spectral indices and band-wise power-law SEFD scaling, and includes RFI and an elevation-dependent SEFD correction fitted from the scan-level data. We applied this model to 25 sessions on the Hobart–Katherine baseline, covering 81 sources over 3–13.5 GHz. Reliable system-temperature (Tsys) measurements were unavailable across these sessions, so the model uses scan-level SNR together with single-session catalogue anchors for each source, comprising an X-band flux-density value and a catalogue-derived S/X spectral index. It recovers flux density and baseline SEFD jointly across all 25 sessions and the fitted channels, each with full posterior uncertainty, and flags instrumental issues and a persistent 4 GHz RFI peak. Agreement with SKED measurements from sessions not used to construct the priors supports the recovered temporal evolution. The hierarchical approach shares information across sources and channels via partial pooling, reducing calibration uncertainty without relying on small reference source sets. This provides a diagnostic tool for detecting instrumental problems and could support SNR-based schedule optimisation, and may also apply to calibration at next-generation facilities such as SKA-Low and DSA-2000.

An individual reports subjective beliefs over continuous events using a proper scoring rule, such as the Quadratic Scoring Rule. Under mild additional assumptions, it is known that these reports reflect latent subjective beliefs if the individual is risk neutral. We demonstrate how to fully recover latent subjective beliefs if the individual is known to distort probabilities into decision weights using Rank Dependent Utility theory, and generalize all results for the complete class of proper scoring rules. As an immediate corollary, we can then exactly recover latent subjective beliefs using Expected Utility Theory. Using Bayesian econometric methods, we demonstrate how to recover belief distributions for individuals, and illustrate these results by recovering belief distributions about the “competitiveness” of individuals and the addictiveness of smoking. These results significantly widen the domain of applicability of proper scoring rules for eliciting latent subjective belief distributions.

When children behave differently from adults in language tasks, it is often unclear if the underlying cause is non-adult-like representation of relevant information, non-adult-like deployment of adult-like representations, or both non-adult-like representations and non-adult-like deployment. We show how computational cognitive modeling can be used to identify which options could lead to specific non-adult-like language behavior, using the case study of Spanish subject pronoun interpretation by typically developing children. In a picture-selection task, children interpret subject pronouns differently from adults; modeling results suggest that both child and adult pronoun interpretation behavior is best captured by inaccuracy somewhere in the pronoun interpretation process, though how exactly children are inaccurate differs from how exactly adults are inaccurate. So, to become adult-like, children need to learn how to be inaccurate in adult-like ways. We discuss the promise and limitations of the computational cognitive modeling approach demonstrated here for evaluating specific hypotheses about the underlying cognitive computations leading to observed language behavior.

Accurate modeling of the loss distributions is crucial for solvency capital assessment and reinsurance pricing. Conventional single-component models fail to capture the dual nature of insurance data, which consists of a high-frequency body and a heavy tails of extreme losses. This paper introduces a novel composite Weibull–Pareto model with a smooth logistic transition in the hazard rates by avoiding discontinuities. We compare frequentist (maximum likelihood) and the Bayesian estimation using weakly informative and empirical priors to assess parameter uncertainty and tail risk. Using the canonical Danish Fire Insurance dataset, the results indicate that Bayesian inference delivers superior predictive performance and conservative estimates of the extreme tail-risk measures such as Value-at-Risk (VaR) and Tail Value-at-Risk (TVaR). Maximum likelihood estimates appear overly optimistic, whereas the Bayesian methods yield more credible capital and reinsurance payouts, particularly for the catastrophic layers. The proposed hazard blend offers a robust framework for actuaries, enhancing both methodological rigor and practical applicability in the composite loss modeling.

The combination of human forecasters’ subjective probability estimates usually improves upon the estimates provided by individual forecasters. In order to combine the probability estimates in a Bayes optimal way, prior work proposed a normative Bayesian fusion model that models the estimates with a beta distribution conditioned on their truth value. However, this model assumes conditionally independent probability estimates, although estimates provided by different forecasters are usually correlated. Here, we introduce a Bayesian model for combining subjective probability estimates that explicitly considers their correlation. We assume that an estimate provided by a forecaster for a given query depends on both the forecaster’s skill and the query’s difficulty. The correlation between probability estimates provided by different forecasters is assumed to be caused by the queries that make the forecasters provide similar estimates, for example, correct and highly confident estimates for very easy queries. Our model represents the probability estimates with a beta distribution conditioned on their truth value. It explicitly models the forecasters’ skills and the queries’ difficulties with skill parameters specific for each forecaster and difficulty parameters specific for each query. In this way, it can model the correlations between probability estimates and consider it when combining the estimates. Evaluations on a data set consisting of the subjective probability estimates of 85 human forecasters for 180 queries show improved fusion performance in terms of Brier score compared to related Bayesian fusion models. In particular, it outperforms independent fusion models that suffer from overconfidence.

This article investigates phonological complexity by using artificial neural network and Bayesian structural equation models to derive representations of phonological complexity from counts of the segments associated with particular features in languages’ phonemic inventories. These latent representations can then be used alongside principal component analysis to further analyse how interactions between phonological features affect overall complexity, and what phonological complexity patterns a model can detect in a phonological feature data set. The results indicate that the per-feature segment counts investigated tend to contribute positively to a language’s complexity, and that the latent complexity variables approximate a log-normal distribution. This implies that phonological complexifications co-occur with other complexifications diachronically while tending to be more constrained at the upper and lower ends of the complexity range.

A key factor in ensuring the accuracy of computer simulations that model physical systems is the proper calibration of their parameters based on real-world observations or experimental data. Bayesian methods provide a robust framework for quantifying and propagating the uncertainties that inevitably arise. Nevertheless, they produce predictions unable to represent the observed datapoints when paired with inexact models. Additionally, the quantified uncertainties of these overconfident models cannot be propagated to other Quantities of Interest (QoIs) reliably. A promising solution involves embedding a model inadequacy term in the inference parameters, allowing the quantified model form uncertainty to influence non-observed QoIs. In this work, we revisit this embedded formulation and analyze how different likelihood constructions affect the inference of model form uncertainty, particularly under the presence of prescribed measurement noise and unavoidable model discrepancies. Two additional likelihood formulations, the global moment-matching and relative global moment-matching likelihoods, are introduced to explore alternative ways of representing the residual distribution. The behavior of these likelihoods is examined alongside existing formulations to show how different treatments of measurement noise and discrepancies shape the inferred parameter posteriors, and thereby affect the uncertainty ultimately propagated to the QoIs. Particular attention is given to how the uncertainty associated with the model inadequacy term propagates to the QoIs for the posteriors obtained from different likelihood formulations, enabling a more comprehensive statistical analysis of the prediction’s reliability. Finally, the proposed approach is applied to estimate the uncertainty in the predicted heat flux from a transient thermal simulation using temperature observations.

We explore the effects of fiscal policy shocks on aggregate output and inflation. We use the Bayesian econometric methodology of Baumeister and Hamilton applied to the fiscal structural vector autoregressive model to evaluate key elasticities and fiscal multipliers using U.S. data. In our baseline specification that ends before Covid pandemic, the government spending multiplier is equal to approximately $0.57$ and tax multiplier is approximately $-0.35$ after one year. The short-term output elasticity of government spending is statistically insignificant and the output elasticity of taxes is approximately equal to $2.26$.

This study presents a framework that combines Bayesian inference with reinforcement learning to guide drone-based sampling for methane source estimation. Synthetic gas concentration and wind observations are generated using a calibrated model derived from real-world drone measurements, providing a more representative testbed that captures atmospheric boundary layer variability. We compare three path planning strategies—preplanned, myopic (short-sighted), and non-myopic (long-term)—and find that non-myopic policies trained via deep reinforcement learning consistently yield more precise and accurate estimates of both source location and emission rate. We further investigate centralized multi-agent collaboration and observe comparable performance to independent agents in the tested single-source scenario. Our results suggest that effective source term estimation depends on correctly identifying the plume and obtaining low-noise concentration measurements within it. Precise localization further requires sampling in close proximity to the source, including slightly upwind. In more complex environments with multiple emission sources, multi-agent systems may offer advantages by enabling individual drones to specialize in tracking distinct plumes. These findings support the development of intelligent, data-driven sampling strategies for drone-based environmental monitoring, with potential applications in climate monitoring, emission inventories, and regulatory compliance.

Approaches to linguistic areas have largely focused either on purely qualitative investigation of area-formation processes, on quantitative and qualitative exploration of synchronic distributions of linguistic features without considering time, or on theoretical issues related to the definition of the notion ‘linguistic area’. What is still missing are approaches that supplement qualitative research on area-formation processes with quantitative methods. Taking a bottom-up approach, we bypass notional issues and propose to quantify area-formation processes by (i) measuring the change in linguistic similarity given a geographical space, a sociocultural setting, a time span, a language sample, and a set of linguistic data, and (ii) testing the tendency and magnitude of the process using Bayesian inference. Applying this approach to the expression of reflexivity in a dense sample of languages in northwestern Europe from the early Middle Ages to the present, we show that the method yields robust quantitative evidence for a substantial gain in linguistic similarity that sets the languages of Britain and Ireland apart from languages spoken outside of Britain and Ireland and cross-cuts lines of linguistic ancestry.

Accurate assessment of adverse event (AE) incidence is critical in clinical research for drug safety. While meta-analysis serves as an essential tool to comprehensively synthesize the evidence across multiple studies, incomplete AE reporting in clinical trials remains a persistent challenge. In particular, AEs occurring below study-specific reporting thresholds are often omitted from publications, leading to left-censored data. Failure to account for these censored AE counts can result in biased AE incidence estimates. We present an R Shiny application that implements a Bayesian meta-analysis model specifically designed to incorporate censored AE data into the estimation process. This interactive tool provides a user-friendly interface for researchers to conduct AE meta-analyses and estimate the AE incidence probability using an unbiased approach. It also enables direct comparisons between models that either incorporate or ignore censoring, highlighting the biases introduced by conventional approaches. This tutorial demonstrates the Shiny application’s functionality through an illustrative example on meta-analysis of PD-1/PD-L1 inhibitor safety and highlights the importance of this tool in improving AE risk assessment. Ultimately, the new Shiny app facilitates more accurate and transparent drug safety evaluations. The Shiny-MAGEC app is available at: https://zihanzhou98.shinyapps.io/Shiny-MAGEC/.

Recent experiments aiming to measure phenomena predicted by strong-field quantum electrodynamics (SFQED) have done so by colliding relativistic electron beams and high-power lasers. In such experiments, measurements of collision parameters are not always feasible. However, precise knowledge of these parameters is required to accurately test SFQED.

Here, we present a novel Bayesian inference procedure that infers collision parameters that could not be measured on-shot. This procedure is applicable to all-optical non-linear Compton scattering experiments investigating radiation reaction. The framework allows multiple diagnostics to be combined self-consistently and facilitates the inclusion of known information pertaining to the collision parameters. Using this Bayesian analysis, the relative validity of the classical, quantum-continuous and quantum-stochastic models of radiation reaction was compared for several test cases, which demonstrates the accuracy and model selection capability of the framework and highlight its robustness if the experimental values of fixed parameters differ from their values in the models.