SEARCH · Searcharxiv

Results for “stat.OT”

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,117 records · Page 62Linked to original sources

arXiv2025

Mapping the Intersection of Research and Policy in Centers for Medicare National Coverage Decision Memos

Evidence is a crucial component of federal policy, but the interactions between the various stakeholders involved in funding, producing, and using the results of scientific research, an important class of evidence, for federal policy are poorly understood. The national coverage determination process used by the Centers for Medicare and Medicaid Services (CMS) to make significant policies on healthcare coverage is an ideal candidate for studying the interactions between stakeholders producing and utilizing scientific research for policy. Memos produced during the national coverage determination process contain information that identifies the organizations funding and producing research articles cited by CMS policy staff. I use these data to map scientific articles and their funding sources to discrete federal policies with substantial economic and health impacts. My analysis highlights that information derived from policy documents can facilitate transparency among the stakeholders involved in funding, producing, and using evidence for federal policy.

econ.GN↗

arXiv2025

Statisticians Training STEM Educators in Statistics Methods and Pedagogy: A Case Study of Instructor Training in Bayesian Methods

Educating the next generation of scientists in statistical methodology is an important task. Educating their instructors in statistical content knowledge and pedagogical knowledge is as important and provides an indirect impact of students' learning. Statisticians are in a place to lead train-the-trainer (TTT) programs in different methods. We present our instructor training program in Bayesian methods as an effective case study of a TTT model. In addition to describing the details of the structure of our training program, we share our experience in designing and implementing our program including the challenges we face, the opportunities created, and our recommendations for TTT programs led by statisticians.

stat.OT↗

arXiv2025

Developing Assessment Methods for Evaluating Learning Experience

This research aims to investigate the gender-based learning experiences of engineering students enrolled in the Probability and Statistics course, focusing on the four different assessment methods employed namely direct conceptual learning (DCL), symposium, applied deployment and collaborative learning. The study encompasses 299 engineering students, comprising 90 females and 209 males. Multivariate Analysis of Variance (MANOVA), is used to gain deeper insights into the complex interplay between assessment methods and their influence on student learning. The results of the statistical analysis reveal that there are significant differences in the learning outcomes between female and male engineering students in the assessment methods of direct conceptual learning, symposium, and applied deployment. The findings suggest that there is no significant difference in the learning outcomes between female and male engineering students in the collaborative learning assessment method. The graphical representation visually confirms the significant differences in direct conceptual learning, symposium, and applied deployment, while illustrating no significant difference in collaborative learning between female and male engineering students.

stat.OT↗

arXiv2026

Bayesian Estimation of Causal Effects Using Proxies of a Latent Interference Network

Network interference occurs when treatments assigned to some units affect the outcomes of others. Traditional approaches often assume that the observed network correctly specifies the interference structure. However, in practice, researchers frequently only have access to proxy measurements of the interference network due to limitations in data collection or potential mismatches between measured networks and actual interference pathways. In this paper, we introduce a framework for estimating causal effects when only proxy networks are available. Our approach leverages a structural causal model that accommodates diverse proxy types, including noisy measurements, multiple data sources, and multilayer networks, and defines causal effects as interventions on population-level treatments. The latent nature of the true interference network poses significant challenges. To overcome them, we develop a Bayesian inference framework. We propose a Block Gibbs sampler with Locally Informed Proposals to update the latent network, thereby efficiently exploring the high-dimensional posterior space composed of both discrete and continuous parameters. The latent network updates are driven by information from the proxy networks, treatments, and outcomes. We illustrate the performance of our method through numerical experiments, demonstrating its accuracy in recovering causal effects even when only proxies of the interference network are available.

stat.ME↗

arXiv2025

Machine Learning Solutions Integrated in an IoT Healthcare Platform for Heart Failure Risk Stratification

The management of chronic Heart Failure (HF) presents significant challenges in modern healthcare, requiring continuous monitoring, early detection of exacerbations, and personalized treatment strategies. In this paper, we present a predictive model founded on Machine Learning (ML) techniques to identify patients at HF risk. This model is an ensemble learning approach, a modified stacking technique, that uses two specialized models leveraging clinical and echocardiographic features and then a meta-model to combine the predictions of these two models. We initially assess the model on a real dataset and the obtained results suggest that it performs well in the stratification of patients at HR risk. Specifically, we obtained high sensitivity (95\%), ensuring that nearly all high-risk patients are identified. As for accuracy, we obtained 84\%, which can be considered moderate in some ML contexts. However, it is acceptable given our priority of identifying patients at risk of HF because they will be asked to participate in the telemonitoring program of the PrediHealth research project on which some of the authors of this paper are working. The initial findings also suggest that ML-based risk stratification models can serve as valuable decision-support tools not only in the PrediHealth project but also for healthcare professionals, aiding in early intervention and personalized patient management. To have a better understanding of the value and of potentiality of our predictive model, we also contrasted its results with those obtained by using three baseline models. The preliminary results indicate that our predictive model outperforms these baselines that flatly consider features, \ie not grouping them in clinical and echocardiographic features.

stat.OT↗

arXiv2025

Basic model for ranking microfinance institutions

This paper discusses the challenges encountered in building a ranking model for aggregator site products, using the example of ranking microfinance institutions (MFIs) based on post-click conversion. We suggest which features of MFIs should be considered, and using an algorithm based on Markov chains, we demonstrate the ``usefulness'' of these features on real data. The ideas developed in this work can be applied to aggregator websites in microinsurance, especially when personal data is unavailable. Since we did not find similar datasets in the public domain, we are publishing our dataset with a detailed description of its attributes.

cs.IR↗

arXiv2025

Bayesian Multivariate Approach to Subnational mortality graduation with Age-Varying Smoothness

This work introduces a Bayesian smoothing approach for the joint graduation of mortality rates across multiple populations. In particular, dynamical linear models are used to induce smoothness across ages through structured dependence, analogously to how temporal correlation is accommodated in state-space time-indexed models. An essential issue in subnational mortality probabilistic modelling is the lack or sparseness of information for some subpopulations. For many countries, mortality data is severely limited, and approaches based on a single population model can result in high uncertainty in the adjusted mortality tables. Here, we recognize the interdependence within a group of mortality data and pursue the pooling of information across several curves that ideally share common characteristics, such as the influence of epidemics or major economic shifts. Our proposal considers multivariate Bayesian dynamical models with common parameters, allowing for borrowing of information across mortality tables and enabling tests of convergence across populations. We also employ discount factors, typical in DLMs, to regulate smoothness, with varying discounting across ages, ensuring less smoothness at younger ages and greater stability at adult ages. This setup implies a trade-off between stability and adaptability. The discount parameter controls the responsiveness of the fit at older ages to new data. The estimation is fully Bayesian, accommodating all uncertainties in modelling and prediction. To illustrate the effectiveness of our model, we analyse male and female mortality data from England and Wales between 2010 and 2012, obtained from the Office for National Statistics. In scenarios with simulated missing data, our approach showed strong performance and flexibility in pooling information from related populations with more complete data.

stat.AP↗

arXiv2025

Sensitivity of ECG QRS Complexes to His-Purkinje Structure in Computational Heart Models

Cardiac digital twins (CDT) are emerging as a potentially transformative tool in cardiology. A critical yet understudied determinant of CDT accuracy is the His-Purkinje system (HPS), which influences ventricular depolarization and shapes the QRS complex of the electrocardiogram (ECG). Here, we quantify how structural variations in the HPS alter QRS morphology and identify which parameters drive this variability. We generated HPS structures using a fractal-tree, rule-based algorithm, systematically varying nine model parameters and assessing their effects on ten QRS-related metrics. We conducted a Sobol sensitivity analysis to quantify direct and interaction-driven contributions of each parameter to observed variability. Our results suggest that most minor changes in HPS structure exert minimal influence on individual QRS features; however, certain parameter combinations can produce abnormal QRS morphologies. Wave durations and peak amplitudes of the QRS complex exhibit low sensitivity to individual HPS parameter variations; however, we found that specific parameter combinations can result in interactions that significantly alter these aspects of QRS morphology. We found that certain HPS structures can cause premature QRS formation, obscuring P-wave formation. QRS timing variability was primarily driven by interactions among branch and fascicle angles and branch repulsivity, though other parameters also showed notable interaction effects. In addition to interactions, individual variations in the number of branches in the HPS also affected QRS timing. While future models should account for these potential sources of variability, this study indicates that minor anatomical differences between a healthy patient's HPS and that of a generic model are unlikely to significantly impact model fidelity or clinical interpretation when both systems are physiologically normal.

q-bio.QM↗

arXiv2025

Larger cities, more commuters, more crime? The role of inter-city commuting in the scaling of urban crime

Cities attract a daily influx of non-resident commuters, reflecting their roles within wider urban networks -- not as isolated places. However, it remains unclear how this interconnectivity shapes the way crime scales with population, given that larger cities tend to receive more commuters and experience more crime. In this work, we investigate how inter-city commuting relates to the population-crime relationship. We find that larger cities receive proportionately more commuters, which in turn is associated with higher levels of burglary, drug possession, robbery, shoplifting, and theft. For example, each 1% increase in inbound commuters corresponds to a 0.32% rise in theft and 0.20% rise in burglary, holding population size constant. We demonstrate that models incorporating both population size and commuter inflows explain variation in these offenses better than population-only models. Our findings underscore the importance of considering how cities are connected -- not just their population size -- in disentangling the population-crime relationship.

physics.soc-ph↗

arXiv2025

Adaptive tail index estimation: minimal assumptions and non-asymptotic guarantees

A notoriously difficult challenge in extreme value theory is the choice of the number $k\ll n$, where $n$ is the total sample size, of extreme data points to consider for inference of tail quantities. Existing theoretical guarantees for adaptive methods typically require second-order assumptions or von Mises assumptions that are difficult to verify and often come with tuning parameters that are challenging to calibrate. This paper revisits the problem of adaptive selection of $k$ for the Hill estimator. Our goal is not an `optimal' $k$ but one that is `good enough', in the sense that we strive for non-asymptotic guarantees that might be sub-optimal but are explicit and require minimal conditions. We propose a transparent adaptive rule that does not require preliminary calibration of constants, inspired by `adaptive validation' developed in high-dimensional statistics. A key feature of our approach is the consideration of a grid for $k$ of size $ \ll n $, which aligns with common practice among practitioners but has remained unexplored in theoretical analysis. Our rule only involves an explicit expression of a variance-type term; in particular, it does not require controlling or estimating a biasterm. Our theoretical analysis is valid for all heavy-tailed distributions, specifically for all regularly varying survival functions. Furthermore, when von Mises conditions hold, our method achieves `almost' minimax optimality with a rate of $\sqrt{\log \log n}~ n^{-|ρ|/(1+2|ρ|)}$ when the grid size is of order $\log n$, in contrast to the $ (\log \log (n)/n)^{|ρ|/(1+2|ρ|)} $ rate in existing work. Our simulations show that our approach performs particularly well for ill-behaved distributions.

stat.OT↗

arXiv2025

Modelling benthic animals in space and time using Bayesian Point Process with cross validation: the case of Holoturians

Understanding the spatial distribution of Holothurians is an essential task for ecosystem monitoring and sustainable management, particularly in the Mediterranean habitats. However, species distribution modeling is often complicated by the presence-only nature of the data and heterogeneous sampling designs. This study develops a spatio-temporal framework based on Log-Gaussian Cox Processes to analyze Holothurians' positions collected across nine survey campaigns conducted from 2022 to 2024 near Giglio Island, Italy. The surveys combined high-resolution photogrammetry with diver-based visual censuses, leading to varying detection probabilities across habitats, especially within Posidonia oceanica meadows. We adopt a model with a shared spatial Gaussian process component to accommodate this complexity, accounting for habitat structure, environmental covariates, and temporal variability. Model estimation is performed using Integrated Nested Laplace Approximation. We evaluate the predictive performances of alternative model specifications through a novel k-fold cross-validation strategy for point processes, using the Continuous Ranked Probability Score. Our approach provides a flexible and computationally efficient framework for integrating heterogeneous presence-only data in marine ecology and comparing the predictive ability of alternative models.

stat.AP↗

arXiv2026

Pivoting the paradigm: the role of spreadsheets in K-12 data science

Spreadsheet tools are widely accessible to and commonly used by K-12 students and teachers. While spreadsheets are not ideal for many types of statistical analysis, they have an important role in data collection and organization. From a pedagogical standpoint, spreadsheets make data visible and easy to interact with, facilitating student engagement in data exploration, analysis, and computation. Though not suitable for all tasks, spreadsheets can facilitate learning and practicing data and computing skills for K-12 students. This paper 1) demonstrates the potential utility of spreadsheets in K-12; 2) reviews prior frameworks and standards that are relevant for K-12 data tools; and 3) proposes data-driven data skills to help develop data acumen and computational fluency. We provide some example activities, identify challenges and barriers to adoption, suggest pedagogical approaches to ease the learning curve for instructors and students, and discuss the need for professional development to facilitate deeper use of spreadsheets for data science and STEM disciplines.

stat.OT↗

arXiv2025

Win Probabilities, Hand Sizes, and Game Duration Analysis in the Bhikar-Sawkar Card Game

We present a Monte Carlo simulation study of the Bhikar-Sawkar card game, a non-deterministic game structurally similar to the classic Beggar-My-Neighbour, which is fully deterministic. Although both games share a common setup, key differences in their rules, particularly the reshuffling of cards after each won hand in Bhikar-Sawkar, introduce stochasticity and significantly increase the space of possible game evolutions. This inherent randomness raises a range of interesting statistical questions regarding the duration of the game, the hand-winner distributions, and the probability of winning the game for a given player. These questions are systematically investigated through large-scale simulations across multiple game configurations.

stat.OT↗

arXiv2025

The cost of ensembling: is it always worth combining?

Given the continuous increase in dataset sizes and the complexity of forecasting models, the trade-off between forecast accuracy and computational cost is emerging as an extremely relevant topic, especially in the context of ensemble learning for time series forecasting. To asses it, we evaluated ten base models and eight ensemble configurations across two large-scale retail datasets (M5 and VN1), considering both point and probabilistic accuracy under varying retraining frequencies. We showed that ensembles consistently improve forecasting performance, particularly in probabilistic settings. However, these gains come at a substantial computational cost, especially for larger, accuracy-driven ensembles. We found that reducing retraining frequency significantly lowers costs, with minimal impact on accuracy, particularly for point forecasts. Moreover, efficiency-driven ensembles offer a strong balance, achieving competitive accuracy with considerably lower costs compared to accuracy-optimized combinations. Most importantly, small ensembles of two or three models are often sufficient to achieve near-optimal results. These findings provide practical guidelines for deploying scalable and cost-efficient forecasting systems, supporting the broader goals of sustainable AI in forecasting. Overall, this work shows that careful ensemble design and retraining strategy selection can yield accurate, robust, and cost-effective forecasts suitable for real-world applications.

cs.LG↗

arXiv2026

Analyzing the retraining frequency of global forecasting models: towards more stable forecasting systems

Forecast stability, that is, the consistency of predictions over time, is essential in business settings where sudden shifts in forecasts can disrupt planning and erode trust in predictive systems. Despite its importance, stability is often overlooked in favor of accuracy. In this study, we evaluate the stability of point and probabilistic forecasts across several retraining scenarios using three large forecastingdatasets and ten different global forecasting models. To analyze stability in the probabilistic setting, we propose a new model-agnostic, distribution-free, and scale-free metric that measuresprobabilistic stability: the Scaled Multi-Quantile Change (SMQC). The results show that less frequent retraining not only preserves but often improves forecast stability, challenging the need for frequent retraining. Moreover, the study shows that accuracy and stability are not necessarily conflicting objectives when adopting a global modeling approach. The study promotes a shift toward stability-aware forecasting practices, proposing a new metric to evaluate forecast stability effectively in probabilistic settings, and offering practical guidelines for building more stable and sustainable forecasting systems.

stat.AP↗

arXiv2025

A Statistical Framework for Model Selection in LSTM Networks

Long Short-Term Memory (LSTM) neural network models have become the cornerstone for sequential data modeling in numerous applications, ranging from natural language processing to time series forecasting. Despite their success, the problem of model selection, including hyperparameter tuning, architecture specification, and regularization choice remains largely heuristic and computationally expensive. In this paper, we propose a unified statistical framework for systematic model selection in LSTM networks. Our framework extends classical model selection ideas, such as information criteria and shrinkage estimation, to sequential neural networks. We define penalized likelihoods adapted to temporal structures, propose a generalized threshold approach for hidden state dynamics, and provide efficient estimation strategies using variational Bayes and approximate marginal likelihood methods. Several biomedical data centric examples demonstrate the flexibility and improved performance of the proposed framework.

stat.ML↗

arXiv2026

One-dimensional quantile-stratified sampling and its application in statistical simulations

In this paper we examine quantile-stratified samples from a known univariate probability distribution, with stratification occurring over a partition of the quantile regions in the distribution. We examine some general properties of this sampling method and we contrast it with standard IID sampling to highlight its similarities and differences. We examine the applications of this sampling method to various statistical simulations including importance sampling. We conduct simulation analysis to compare the performance of standard importance sampling against the quantile-stratified importance sampling to see how they each perform on a range of functions.

stat.ME↗

arXiv2025

Next-Generation Conflict Forecasting: Unleashing Predictive Patterns through Spatiotemporal Learning

Forecasting violent conflict at high spatial and temporal resolution remains a central challenge for both researchers and policymakers. This study presents a novel neural network architecture for forecasting three distinct types of violence -- state-based, non-state, and one-sided -- at the subnational (priogrid-month) level, up to 36 months in advance. The model jointly performs classification and regression tasks, producing both probabilistic estimates and expected magnitudes of future events. It achieves state-of-the-art performance across all tasks and generates approximate predictive posterior distributions to quantify forecast uncertainty. The architecture is built on a Monte Carlo Dropout Long Short-Term Memory (LSTM) U-Net, integrating convolutional layers to capture spatial dependencies with recurrent structures to model temporal dynamics. Unlike many existing approaches, it requires no manual feature engineering and relies solely on historical conflict data. This design enables the model to autonomously learn complex spatiotemporal patterns underlying violent conflict. Beyond achieving state-of-the-art predictive performance, the model is also highly extensible: it can readily integrate additional data sources and jointly forecast auxiliary variables. These capabilities make it a promising tool for early warning systems, humanitarian response planning, and evidence-based peacebuilding initiatives.

stat.OT↗