Review
Mathematical & Computational Biology
Julie M. Petersen, Malcolm Barrett, Katherine A. Ahrens, Eleanor J. Murray, Allison S. Bryant, Carol J. Hogue, Sunni L. Mumford, Salini Gadupudi, Matthew P. Fox, Ludovic Trinquart
Summary: Systematic reviews and meta-analyses are crucial for determining the etiological associations between exposures or interventions and health outcomes. However, observational studies are susceptible to residual confounding, which can affect their validity. The confounder matrix approach provides a method for assessing and summarizing the control of confounding in observational studies, improving the transparency of reporting and informing meta-analyses.
RESEARCH SYNTHESIS METHODS
(2022)
Article
Computer Science, Information Systems
Ninghui Li, Xiaokuan Zhang, Binfeng Zong, Fan Lv, Jiahua Xu, Zhaolong Wang
Summary: The DOA estimation of wideband signals based on sparse signal reconstruction has been proposed due to its high-resolution performance. To enhance sparsity, a novel hierarchical Bayesian prior framework and an iterative approach were proposed, which showed lower computational complexity than current state-of-the-art algorithms. The proposed approach achieved high angular estimation accuracy and sparsity performance by utilizing the joint sparsity of MMV models and stabilizing the estimated values between different frequencies or snapshots to obtain a flat spatial spectrum. Extensive simulation results demonstrated the superior performance of the method.
Review
Medicine, General & Internal
Stan R. W. Wijn, Maroeska M. Rovers, Gerjon Hannink
Summary: Researchers commonly use propensity score matching (PSM) to adjust for confounding in longitudinal observational data, but inappropriate use of PSM was found in 25% of studies with a time-varying treatment. While methods designed for time-varying treatment and confounding are available, they were only utilized in 45% of the studies.
Article
Automation & Control Systems
Xiaolong Chen, Yi Chai, Qie Liu, Pengfei Huang, Linchuan Fan
Summary: In this paper, a novel Bayesian sparse multiple kernel-based identification method (BSMKM) for multiple-input single-output (MISO) Hammerstein system is proposed. The method represents the nonlinear part and the linear part using basis-function model and finite impulse response model respectively and estimates all unknown model parameters through hierarchical prior distribution and full Bayesian method based on variational Bayesian inference.
Article
Multidisciplinary Sciences
William J. Harrison, Paul M. Bays, Reuben Rideaux
Summary: Perception is often modeled as a process of active inference, with prior expectations combined with sensory measurements to estimate the world's structure. The authors use a data-driven approach to extract the brain's representation of visual orientation and compare it with different sensory coding schemes. They find that the tuning of the human visual system is highly conditional on stimulus-specific variations, and the adopted encoding scheme embeds an environmental prior for optimal inference in cortical processing.
NATURE COMMUNICATIONS
(2023)
Review
Cardiac & Cardiovascular Systems
Bart J. J. Velders, J. W. Taco Boltje, Michiel D. Vriesendorp, Robert J. M. Klautz, Saskia Le Cessie, Rolf H. H. Groenwold
Summary: This study systematically evaluated the quality of conduct and reporting of confounding adjustment methods in observational studies on cardiothoracic interventions. The findings showed insufficient reporting of these methods, making it difficult to assess the quality. Proper application of confounding adjustment methods is crucial for causal inference on optimal treatment strategies in clinical practice.
EUROPEAN JOURNAL OF CARDIO-THORACIC SURGERY
(2023)
Article
Social Sciences, Mathematical Methods
Jouni Helske, Santtu Tikka, Juha Karvanen
Summary: This study investigates how to estimate causal effects when well-known adjustments are not applicable. It finds that biases in small samples may arise when a causal effect is subject to an implicit functional constraint, related to trapdoor variables. By studying different strategies to account for trapdoor variables and minimizing the bias, a more accurate estimation of causal effects can be achieved. The importance of trapdoor variables in causal effect estimation is illustrated using real data from a specific study, with Bayesian modeling allowing for parameter uncertainty to be taken into account.
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-STATISTICS IN SOCIETY
(2021)
Article
Statistics & Probability
Widemberg S. Nobre, Alexandra M. Schmidt, Joao B. M. Pereira
Summary: The article addresses the issue of spatial confounding in spatial generalized linear models, particularly in multilevel spatial models with multiple observations within clusters. It explores the potential bias of cluster-level fixed effects and proposes restricted spatial regression as a remedy. Additionally, the paper briefly touches on the issue of confounding in random intercept and slope models.
INTERNATIONAL STATISTICAL REVIEW
(2021)
Article
Mathematical & Computational Biology
Maura Mezzetti, Colleen P. Ryan, Priscilla Balestrucci, Francesco Lacquaniti, Alessandro Moscatelli
Summary: The aim of this article is to reconsider hierarchical models in a Bayesian framework, which have seldom been applied in the analysis of psychometric functions. The main advantage of using Bayesian models is the reduction of parameter uncertainty through the combination of prior knowledge and experimental data. The Bayesian hierarchical model, implemented using JAGS and rjags, provides a promising and powerful method for analyzing psychometric functions.
FRONTIERS IN COMPUTATIONAL NEUROSCIENCE
(2023)
Article
Physics, Multidisciplinary
Valeria Leiva-Yamaguchi, Danilo Alvares
Summary: Models that combine longitudinal and survival outcomes have become popular, but the inferential process can be time-consuming. To reduce complexity while maintaining accuracy, a two-stage strategy is proposed where the longitudinal submodel is first fitted, followed by plugging shared information into the survival submodel.
Review
Mathematical & Computational Biology
Anita Natalia Varga, Alejandra Elizabeth Guevara Morel, Joran Lokkerbol, Johanna Maria van Dongen, Maurits Willem van Tulder, Judith Ekkina Bosmans
Summary: The aim of this article is to review the methods available for dealing with confounding in analyzing the effect of health care treatments with single-point exposure in observational data. The results show that there are significant differences in performance between different methods, and the performance of a specific method is highly dependent on the estimator used.
STATISTICS IN MEDICINE
(2023)
Article
Computer Science, Interdisciplinary Applications
Nikolas Kuschnig, Lukas Vashold
Summary: This paper introduces BVAR, an R package dedicated to estimating Bayesian VAR models with hierarchical prior selection, providing functionalities for a wide range of research problems. The package offers a user-friendly and transparent interface, making Bayesian VAR models accessible for users.
JOURNAL OF STATISTICAL SOFTWARE
(2021)
Article
Mathematical & Computational Biology
Ethan M. Alt, Matthew A. Psioda, Joseph G. Ibrahim
Summary: There is growing interest in using prior information in statistical analyses. In rare diseases, it can be challenging to establish treatment efficacy solely based on prospective study data due to small sample sizes. To address this issue, an informative prior can be elicited to determine treatment effects. This study develops a novel hierarchical prediction prior (HPP) that allows practitioners to elicit a prior prediction for the mean response in generalized linear models. The HPP approach shows higher efficiency gains compared to the conjugate prior and the power prior when predictions are incompatible with the data.
Article
Biochemistry & Molecular Biology
Ben Galvin, Jay Jones, Michaela Powell, Katherine Olin, Matthew Jones, Thomas Robbins
Summary: This research proposes a method to expand the taxonomic resolution of PCR diagnostic systems for pathogen identification by leveraging known genetic variations and post-PCR melting curve analysis. The approach can be used to monitor outbreaks, observe circulation patterns, and guide testing practices.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2022)
Article
Biology
Chiara Piazzola, Lorenzo Tamellini, Raul Tempone
Summary: This article provides an overview of methods for prediction under uncertainty and data fitting of dynamical systems, with a focus on SIR-like models used in predicting the trend of the COVID-19 pandemic. It warns about the identifiability of parameters in SIR-like models, making it challenging to use these models for accurate predictions. Many of the challenges discussed are generally valid for inverse problems in broader contexts.
MATHEMATICAL BIOSCIENCES
(2021)