Article
Engineering, Mechanical
Jia-Hua Yang, Wen-Yue Liu, Yong-Hui An, Heung-Fai Lam
Summary: This paper develops an enhanced adaptive sequential Monte Carlo (ASMC) method to solve Bayesian model updating and model class selection for complex engineering structures. The study reveals the difficulties of sampling from complex probability density functions (PDFs) during sequential sampling process, leading to the approximation of PDF using incremental weights of samples and the adaptive sampling scheme. The research also presents new formulations to calculate model class evidence, enabling the separate quantification of data fit and information gain of a model class.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2024)
Article
Mathematical & Computational Biology
Dongyu Li, Lei Wang, Weihua Zhao
Summary: This article investigates parameter estimation, kink points testing, and statistical inference for a longitudinal multikink expectile regression model. Estimators for kink locations and regression coefficients are obtained using a bootstrap restarting iterative algorithm to avoid local minima. The study shows the consistency in kink points selection and asymptotic normality of all estimators.
STATISTICS IN MEDICINE
(2022)
Article
Mathematics
Lili Yue, Jianhong Shi, Jingxuan Luo, Jinguan Lin
Summary: This paper considers a one-way error component regression model with measurement errors. The unknown parameter vector is estimated using the bias-corrected method, and its asymptotic properties are developed. For hypothesis testing of the coefficient parameter vector, a parametric bootstrap (PB) method is proposed. The effectiveness of the proposed PB test method is discussed through numerical simulations and real data analysis under different sample sizes and parameter configurations.
Article
Psychology, Biological
Matteo Lisi, Gianluigi Mongillo, Georgia Milne, Tessa Dekker, Andrei Gorea
Summary: The study found that human confidence judgments tend to follow discrete confidence levels rather than Bayesian probabilities. While humans can express confidence about uncertain events, they do not fully adhere to the ideal Bayesian strategy. By developing a dual-decision task, researchers suggest that confidence judgments may be based on point estimates of relevant variables.
NATURE HUMAN BEHAVIOUR
(2021)
Review
Psychology, Mathematical
Rianne de Heide, Peter D. Grunwald
Summary: This article addresses the impact of optional stopping on Bayesian methods, specifying the potential issues under certain circumstances. By extending Rouder's experiments, the authors demonstrate that when parameters are equipped with default or pragmatic priors in practical applications, resilience to optional stopping can be compromised. The article distinguishes between three types of default priors, each with specific issues related to optional stopping.
PSYCHONOMIC BULLETIN & REVIEW
(2021)
Article
Thermodynamics
Yingqi Chen, Shusong Ba, Qing Yang, Tian Yuan, Haibo Zhao, Ming Zhou, Pietro Bartocci, Francesco Fantozzi
Summary: This study examines the market efficiency of the Hubei pilot carbon market and finds a significant increase in market efficiency post-COVID-19. The findings provide important insights for analyzing the impact of the pandemic on the industrial sector.
Article
Mathematics, Interdisciplinary Applications
Jin Wang, Yunbo Ouyang, Yuan Ji, Feng Liang
Summary: In this study, we explore the Bayesian approach to variable selection in linear regression models. We propose an efficient EM algorithm that returns the MAP estimator of the relevant variables set. The algorithm avoids the need for inverting large matrices in each iteration, making it scalable for big data. Additionally, we introduce an ensemble EM algorithm to address the issue of local modes and achieve better variable selection results. Empirical studies have shown the superior performance of the ensemble EM algorithm.
Article
Engineering, Marine
Guo-Hai Dong, Hui-Min Hou, Tiao-Jian Xu
Summary: This study estimated the model uncertainty in hydrodynamic characteristics by existing numerical models and effectively predicted the uncertainty of the model bias factor using the bootstrap method. The results show that model uncertainty can significantly affect the reliability of the mooring line, and bootstrap method can provide a more reasonable assessment of the failure probability.
Article
Statistics & Probability
Danijel Kivaranovic, Hannes Leeb
Summary: Research on valid inference after model selection is actively exploring the polyhedral method for constructing confidence intervals, with varying lengths depending on the model and computation complexity. Simulation results suggest that the sufficient condition for infinite length is met unless the selected model includes almost all or almost none of the available regressors. Additionally, kappa-quantiles exhibit a similar behavior for kappa close to 1 in the distribution of confidence interval lengths.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2021)
Article
Engineering, Multidisciplinary
Alan M. Polansky, Duchwan Ryu
Summary: The structure of a system is crucial in determining the relationship between component reliability and system reliability. Calculating system reliability is straightforward under the assumption of independence, but becomes more complex in the case of dependence. Established methods use known system structure and reliability data to assess the independence of individual components.
QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL
(2021)
Article
Environmental Sciences
Maria Fernanda Morales Oreamuno, Sergey Oladyshkin, Wolfgang Nowak
Summary: Bayesian model selection (BMS) and Bayesian model justifiability analysis (BMJ) provide a statistically rigorous framework for comparing competing models using Bayesian model evidence (BME). However, BME-based analysis has limitations in accounting for a model's predictive performance after calibration and in comparing models using different calibration subsets. To address these limitations, we propose augmenting BMS and BMJ analyses with information-theoretic measures such as expected log-predictive density (ELPD), relative entropy (RE), and information entropy (IE). We demonstrate how these measures, alongside BME, enhance the understanding of the Bayesian updating process and enable objective model comparison using different calibration datasets.
WATER RESOURCES RESEARCH
(2023)
Article
Health Care Sciences & Services
Riko Kelter
Summary: The success of preclinical research relies on both exploratory and confirmatory animal studies. Null hypothesis significance testing is commonly used to select the most promising treatments for clinical research, while considering the balance between false discoveries and false omissions. In this paper, we compare different preclinical research pipelines, based on either null hypothesis significance testing or Bayesian statistical decision criteria, and show that incorporating the minimum clinically important difference and using Bayesian statistical decision criteria can improve the reliability of preclinical animal research by reducing false-positive findings.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2023)
Article
Environmental Sciences
Roohollah Noori, Ali Mirchi, Farhad Hooshyaripor, Rabin Bhattarai, Ali Torabi Haghighi, Bjarn Klove
Summary: This study introduces a new index called bandwidths similarity factor (bws-factor) to quantify the reliability of functional forms (FFs) to calculate longitudinal dispersion coefficient (K-x). The results show that the poor reliability of FFs for K-x calculation is mainly due to different sources of error in the K-x calculation process.
SCIENCE OF THE TOTAL ENVIRONMENT
(2021)
Article
Statistics & Probability
Christopher T. Franck, Michael L. Madigan, Nicole A. Lazar
Summary: This paper discusses the controversies surrounding classical hypothesis testing and p values, and proposes principles for describing alternative methods in venues beyond the statistical literature. It also presents an existing BIC-based approximation to Bayesian model selection as a complete alternative approach to classical hypothesis testing.
Article
Geochemistry & Geophysics
Ling Yang, Yunzhong Shen, Bofeng Li, Chris Rizos
Summary: The study presents quality control indices based on the DIA estimators and a simplified algebraic estimation (SAE) method for single outlier detection. By taking into account the uncertainty of the combined estimation-testing procedure, the accuracy and computational efficiency of quality control are improved.
JOURNAL OF GEODESY
(2021)