iiinova — where your digital self keeps evolving

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 23/04/2026

Applied to Norwegian LFS data, the new estimator provides substantial precision gains for estimates of change over time in employment.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 01/04/2026

This framework provides more statistically rigorous and robust measurements, uncertainty intervals, and better characterisation of conflict risk

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 01/04/2026

Many current methods for monitoring conflict risk rely on historical averages, which can be unstable and sensitive to outliers. The authors propose Bayesian,spatiotemporal discrete-time Hawkes processes to support improved monitoring and decision-making

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 30/03/2026

Estimated solely on aggregate data, our model successfully predicts out-of-sample micro outcomes like age at first sex and birth intervals

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 21/11/2025

We demonstrate that higher variability in snap timing is beneficial for the passing game, as it relates to facing less havoc created by the opposing defence. We also obtain a quarterback leaderboard based on our snap timing variability measure, and Patrick Mahomes stands out 🐐

021

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 03/10/2025

High recidivism risk found for 20% of ever-jailed. 10% cycle back to jail up to twice per year. Little association to age, gender, crime type and race.

010

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 03/10/2025

They also illustrate the application of the proposed tree-based MI method using data from a cellphone survey on COVID-19 vaccination in Uganda, which represents a subcohort sample drawn from the 2020 Uganda Population-based HIV Impact Assessment Survey.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 01/10/2025

Fraud detection in blockchain networks presents unique challenges due to decentralized and pseudonymous nature of transactions. This study introduces a novel Multilayer Topology-Aware Graph Contrastive Learning (MTGCL) framework to detect fraudulent activity within the Ethereum transaction network

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 23/09/2025

Their results suggest that overlooking latent homophily can lead to either underestimation or overestimation of causal peer influence, accompanied by considerable estimation uncertainty.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 19/09/2025

The temporal evolution of transmission rates in populations containing multiple types of individual is reconstructed via an appropriate dimension-reduction formulation driven by independent diffusion processes.

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 19/09/2025

Authors propose practical guidelines, and present the performance of the proposed estimators in numerical studies in two sets of real data: exit polls from the 19th South Korean election and public data collected from the Korean Survey of Household Finances and Living Conditions

010

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 03/09/2025

They use a unique form of resampling for valid estimates of our test statistic's null distribution even under violations of standard assumptions. This GeoRDD procedure gives substantially different results in the analysis of NYC arrest rates than those that rely on standard assumptions.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 03/09/2025

They study variation in policing outcomes attributable to differential policing practices in NYC using geographic regression discontinuity designs (GeoRDDs).

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 26/08/2025

Who increases ED use after Medicaid? New causal ML methods reveal effect a small share of Oregon Medicaid experiment recipients drive overall ED use increase, masking wide variation.

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 21/08/2025

When accounting for changepoints, 27.56% of counties had their trend estimates change by >0.03 ppm. While overall ozone levels dropped (thanks to air quality policies), extreme ozone trends actually increased in 45.82% of counties after adjusting for data disruptions.

120

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 21/08/2025

How to accurately measure long-term ozone trends when data keeps getting disrupted? Changes in air quality policies, monitor locations, instruments, and sampling methods create "changepoints" that can make trend analysis misleading if you don't account for them properly.

110

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 21/08/2025

Authors found that the original HES relied too heavily on military security scores. The Bayesian approach showed that giving more weight to socioeconomic factors (not just military metrics) could have led to better decisions while maintaining safety constraints for civilians

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 21/08/2025

The authors developed "Bayesian safe policy learning" - a method that ensures new algorithms won't make things worse for vulnerable groups. They introduced "ACRisk" (% of groups that would be harmed by a policy change) and created safeguards to keep this risk low

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 21/08/2025

How to safely deploy data-driven decision algorithms in high-stakes situations? During Vietnam, the US military used the Hamlet Evaluation System (HES) to score regional security and guide airstrike decisions. But what if this algorithm could have been improved to reduce harm?

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 12/08/2025

The method demonstrates robust performance across various sample sizes and censoring scenarios, including right censoring and missing at random, and results on real world data provide an interpretable and reliable tool for optimizing survival outcomes in complex clinical settings

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 12/08/2025

To effectively manage censored survival data while optimizing treatment strategies the authors propose the Linear Buckley–James Q-Learning framework (BJ-Q), which integrates the Buckley–James method and apply it through extensive simulation studies and real world data

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 12/08/2025

Survival analysis is a fundamental aspect of medical research. Despite its importance, the frequent occurrence of censoring complicates both the analysis and interpretation of survival data. Q-Learning method, despite its promise, presents unique challenges in survival analysis.

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 06/08/2025

We demonstrate how conflict is triggered across cells, varying distances and time lags. We find that diffusion is driven by population structures. Conflict generally breaks out in densely populated areas and from there diffuses across the region and disproportionately affecting less populated areas

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 06/08/2025

We develop a GAM model that smooths across the spatio-temporal history of each observation, to capture these diffusion mechanisms. Using gridded conflict data from UCDP in Africa, we analyze diffusion up to 550km in distance and 24 months in the past.

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 06/08/2025

Armed conflict exhibits substantial diffusion, i.e., spreading, across space and time. However, existing statistical models do not analyze nor fully capture these complex transmission mechanisms.

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 22/07/2025

The findings do not support a momentum effect and, moreover, show that the observed momentum in the data can be explained by differences in players’ strength. Hence, the empirical evidence suggests that past successes per se do not affect the probability of a current success.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 22/07/2025

Results suggest differences in the effects of various vocal interventions, and that average pitch of a speaker’s voice does not necessarily have the largest effect on listeners, highlighting the need for further study of the channels through which speech influences listeners

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 22/07/2025

Results show that the new model can produce far more accurate point and interval estimates, compared to standard approaches and approaches that use shrinkage or spatial priors alone. The model also had reduced uncertainty in the data analysis compared to the direct estimate.

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 30/06/2025

The model is applied to age-specific immigration flows to Austria, disaggregated by sex and countries of origin. Comparative analysis demonstrates that the model outperforms commonly used benchmark frameworks in both in-sample imputation and out-of-sample predictive exercises.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 30/06/2025

Authors conduct a simulation study to investigate the statistical properties of the proposed methods. Finally, they apply the methods to aerial survey data of oil and gas facilities in British Columbia, Canada, to estimate the methane emissions in the province.

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 30/06/2025

Results provide a detailed analysis of high-risk areas for recurrent crimes and the behaviour of recidivism rates over time in Minas Gerais, Brazil. This research significantly enhances our understanding of criminal trajectories, which might help in combating criminal recidivism.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 30/06/2025

The proposed new models are designed for recurrent events data characterized by an excess of zeros and spatial correlation. In addition to their parametric counterparts, we propose flexible semi-parametric versions approximating the intensity function using Bernstein Polynomials

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 17/06/2025

The hidden Markov development model is found to perform comparably to, and frequently better than, the two-step approach, as well as a latent change-point model, on numerical examples and industry datasets.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 17/06/2025

Temperature-related features are most influential in explaining mortality deviations from the baseline over short time periods. Furthermore, we find that environmental features prove particularly beneficial in southern regions for explaining elevated levels of mortality

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 09/06/2025

Our analysis of CRFs in South Carolina reveals that certain counties, such as Chesterfield and Clarendon, exhibit gaps in racial health disparities, making them prime candidates for community-level interventions aimed at reducing these disparities.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 09/06/2025

Authors' interest is in approaches to measuring disparities that provide both graphical displays and summary measures quantifying the extent of disparity. This is why recent modifications of the Lorenz curve referred to as transformed Lorenz curves, allow that

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 03/06/2025

Through simulation studies, authors show that their proposed approach can yield more precise point and interval estimates when compared to either traditional direct estimators, or existing model-based estimators that lack a longitudinal or temporal component.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 03/06/2025

Contrary to what traditional methods find, authors's approach estimates a controlled direct effect of perspective-taking conversations when subjective feelings are neutral but not positive or negative, and this result is robust to moderate departures from parallel trends.

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 05/05/2025

The model permits estimating some‘nowcast’ quantities beyond reproduction numbers. In particular, we have shown how a snapshot of the susceptibility profile of the population can be obtained, indicating that by March 2023 almost all of the population have had a previous infection

000

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 05/05/2025

Extending previous work (Birrell et al., 2021), here authors detail a Bayesian approach to inference that permitted the timely estimation of latent features of the pandemic over the full three years from March 2020 to March 2023

100

Journal of the Royal Statistical Society: Series A @jrssa.bsky.social · 05/05/2025

Over the following three years, continual adaptation and extension were required to deal with a long-lived pandemic subject to unprecedented levels of public health intervention and surveillance.

100