Bayesian Testing Research Articles

5 Antihypertensive Medication Use and Cognition in Older Adults

Objective:Hypertension is a common disorder that has been inconsistently associated with worse cognition in older adults. Antihypertensive medications offer treatment for high blood pressure but previous studies on the association between blood pressure, antihypertensive use, and cognitive performance in older adults have yielded inconsistent findings. Individuals without high blood pressure may also take antihypertensive medications for other medical conditions, including migraines. It is unclear whether antihypertensive medications have any effect on cognitive performance in older adults, and whether the differences, if any, are similar in hypertensives and normotensives.Participants and Methods:4,969 participants from the National Alzheimer Coordinating Center (NACC) database were included in this study (Mage=72.4 years, SD=7.3 years). Cognitive assessment included Letter Fluency, Category Fluency (animals and vegetables), Trail Making Test A & B, Digit Ordering (forward and backward), and MoCA total score. Participants were included if they had a clinician diagnosis of hypertension or normotension and recorded history of whether they take any antihypertensive medication. Participants with a history of stroke were excluded. Cognitive differences between medication groups were investigated in hypertensive participants and normotensive participants using Bayesian Mann-Whitney tests.Results:Bayesian Mann-Whitney tests in hypertensive individuals showed no cognitive differences between those taking antihypertensive medications and those not taking antihypertensives (all BF10s < 3). Bayesian Mann-Whitney tests in normotensive individuals showed individuals taking antihypertensive medications performed worse on Trail Making Test B compared to individuals not taking antihypertensives (123.6 seconds vs 108.8 seconds; BF10 = 35.1), with a small effect size (d=-.156).Conclusions:These results suggest that antihypertensive use in older adults with normal blood pressure may be associated with worse executive functioning. Antihypertensive use in normotensive older adults may lower blood pressure and reduce cerebral perfusion, resulting in worse cognitive functioning. Future studies should investigate long-term antihypertensive use and associated cognitive changes in both hypertensive and normotensive individuals.

Gaussian Boson Sampling with Pseudo-Photon-Number-Resolving Detectors and Quantum Computational Advantage.

We report new Gaussian boson sampling experiments with pseudo-photon-number-resolving detection, which register up to 255 photon-click events. We consider partial photon distinguishability and develop a more complete model for the characterization of the noisy Gaussian boson sampling. In the quantum computational advantage regime, we use Bayesian tests and correlation function analysis to validate the samples against all current classical spoofing mockups. Estimating with the best classical algorithms to date, generating a single ideal sample from the same distribution on the supercomputer Frontier would take ∼600 yr using exact methods, whereas our quantum computer, Jiǔzhāng 3.0, takes only 1.27 μs to produce a sample. Generating the hardest sample from the experiment using an exact algorithm would take Frontier∼3.1×10^{10} yr.

Genome-wide search algorithms for identifying dynamic gene co-expression via Bayesian variable selection.

A wealth of gene expression data generated by high-throughput techniques provides exciting opportunities for studying gene-gene interactions systematically. Gene-gene interactions in a biological system are tightly regulated and are often highly dynamic. The interactions can change flexibly under various internal cellular signals or external stimuli. Previous studies have developed statistical methods to examine these dynamic changes in gene-gene interactions. However, due to the massive number of possible gene combinations that need to be considered in a typical genomic dataset, intensive computation is a common challenge for exploring gene-gene interactions. On the other hand, oftentimes only a small proportion of gene combinations exhibit dynamic co-expression changes. To solve this problem, we propose Bayesian variable selection approaches based on spike-and-slab priors. The proposed algorithms reduce the computational intensity by focusing on identifying subsets of promising gene combinations in the search space. We also adopt a Bayesian multiple hypothesis testing procedure to identify strong dynamic gene co-expression changes. Simulation studies are performed to compare the proposed approaches with existing exhaustive search heuristics. We demonstrate the implementation of our proposed approach to study the association between gene co-expression patterns and overall survival using the RNA-sequencing dataset from The Cancer Genome Atlas breast cancer BRCA-US project.

Semi-supervised clustering with two types of background knowledge: Fusing pairwise constraints and monotonicity constraints

This study addresses the problem of performing clustering in the presence of two types of background knowledge: pairwise constraints and monotonicity constraints. Pairwise constraint relate pair of instances, indicating if they must or must not belong to the same cluster, while monotonicity constraints impose and order relationship between the cluster labels, and are applied to the dataset globally. To fuse these to types of background knowledge, the formal framework to perform clustering under monotonicity constraints is, firstly, defined, resulting in a specific distance measure. Pairwise constraints are integrated afterwards by designing an objective function which combines the proposed distance measure and a pairwise constraint-based penalty term, in order to fuse both types of information. This objective function can be optimized with an EM optimization scheme. The proposed method serves as the first approach to the problem it addresses, as it is the first method designed to work with the two types of background knowledge mentioned above. Our proposal is tested with previous approaches from different clustering paradigms in monotonic datasets, including a real-world case of study. Bayesian statistical tests are used to provide empirical evidence in favor of our proposal, and fund subsequent conclusions: the proposed method can be used obtain better results than previous approaches within the monotonic constrained clustering paradigm.

Ancestral Origin and Dissemination Dynamics of Reemerging Toxigenic Vibrio cholerae, Haiti.

The 2010 cholera epidemic in Haiti was thought to have ended in 2019, and the Prime Minister of Haiti declared the country cholera-free in February 2022. On September 25, 2022, cholera cases were again identified in Port-au-Prince. We compared genomic data from 42 clinical Vibrio cholerae strains from 2022 with data from 327 other strains from Haiti and 1,824 strains collected worldwide. The 2022 isolates were homogeneous and closely related to clinical and environmental strains circulating in Haiti during 2012-2019. Bayesian hypothesis testing indicated that the 2022 clinical isolates shared their most recent common ancestor with an environmental lineage circulating in Haiti in July 2018. Our findings strongly suggest that toxigenic V. cholerae O1 can persist for years in aquatic environmental reservoirs and ignite new outbreaks. These results highlight the urgent need for improved public health infrastructure and possible periodic vaccination campaigns to maintain population immunity against V. cholerae.

The EDIBLES survey

Context. Interstellar absorption observed toward stellar targets changes slowly over long timescales, mainly due to the proper motion of the background target relative to the intervening clouds, such that over time, different parts of the intervening cloud are probed. On longer timescales, the slowly changing physical and chemical conditions in the cloud can also cause variation. Detecting such time variations thus provides an opportunity to study cloud structure. Aims. We searched for systematic variations in the absorption profiles of the diffuse interstellar bands (DIBs) and interstellar atomic and molecular lines by comparing the high-quality data set from the recent ESO diffuse interstellar bands large exploration survey (EDIBLES) to older archival observations, bridging typical timescales of ~10 yr with a maximum timescale of 22 yr. Methods. For 64 EDIBLES targets, we found adequate archival observations. We selected 31 strong DIBs, seven atomic lines, and five molecular lines to focus our search on. We carefully considered various systematic effects and used a robust Bayesian quantitative test to establish which of these absorption features could display significant variations. Results. While systematic effects greatly complicate our search, we find evidence for variations in the profiles of the λλ4727 and 5780 DIBs in a few sightlines. Toward HD 167264, we find a new Ca I cloud component that appears and becomes stronger after 2008. The same sightline furthermore displays marginal, but systematic changes in the column densities of the atomic lines originating from the main cloud component in the sightline. Similar variations are seen toward HD 147933. Conclusions. Our high-quality spectroscopic observations in combination with archival data show that it is possible to probe interstellar time variations on time scales of typically a decade. Despite the fact that systematic uncertainties as well as the generally somewhat lower quality of older data complicate matters, we can conclude that time variations can be made visible, both in atomic lines and DIB profiles for a few targets, but that generally, these features are stable along many lines of sight. We present this study as an archival baseline for future comparisons, bridging longer periods.

Testing higher and infinite degrees of stochastic dominance for small samples: A Bayesian approach

This study proposes a distribution-free Bayesian procedure that detects infinite degrees of stochastic dominance (SD∞) between two random outcomes and then seeks a finite degree k≥1 of stochastic dominance (SDk). Based on small samples, we construct four-choice Bayesian tests by combining an encompassing prior Bayesian model with the Dirichlet process priors that discriminate between SD∞ and SDk of one random variable over the other with non-dominance or equality between them. We use Monte Carlo simulations to evaluate the novel Bayesian tests for SDk and SD∞ and compare them to the subsampling and bootstrap significance tests for SDk. Our simulation shows that the Bayesian tests for SDk outperform the significance tests for small samples, especially for detecting non-stochastic dominance. This study shows that the test for SD∞ is an accurate decision-making tool when using small samples.

Investigation of half-normal model using informative priors under Bayesian structure

This paper considers properties of half-normal distribution using informative priors under the Bayesian criterion. It employs the squared root inverted gamma, Chi-square and Rayleigh distributions as the prior distribution to construct the Posterior distributions of the respective distributional parameters. Hyperparameters are elicited via prior predictive distribution. The properties of posterior distribution are studied, and their graphs are presented using a real data set. A comprehensive simulation scheme is conducted using informative priors. Bayes estimates are obtained using the loss functions (squared error loss function, modified loss function, quadratic loss function and Degroot loss function). Statistical inferences interval estimates and Bayesian hypothesis testing are presented to demonstrate the usefulness of the study.

Bayesian evidence synthesis for informative hypotheses: An introduction.

To establish a theory one needs cleverly designed and well-executed studies with appropriate and correctly interpreted statistical analyses. Equally important, one also needs replications of such studies and a way to combine the results of several replications into an accumulated state of knowledge. An approach that provides an appropriate and powerful analysis for studies targeting prespecified theories is the use of Bayesian informative hypothesis testing. An additional advantage of the use of this Bayesian approach is that combining the results from multiple studies is straightforward. In this article, we discuss the behavior of Bayes factors in the context of evaluating informative hypotheses with multiple studies. By using simple models and (partly) analytical solutions, we introduce and evaluate Bayesian evidence synthesis (BES) and compare its results to Bayesian sequential updating. By doing so, we clarify how different replications or updating questions can be evaluated. In addition, we illustrate BES with two simulations, in which multiple studies are generated to resemble conceptual replications. The studies in these simulations are too heterogeneous to be aggregated with conventional research synthesis methods. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Decision Theoretic Cutoff and ROC Analysis for Bayesian Optimal Group Testing

Decision Theoretic Cutoff and ROC Analysis for Bayesian Optimal Group Testing

Cognitive Neuroimaging Studies on Poverty and Socioeconomic Status Differences in Children and Families across the World: Translational Insights for Next Decade’s Policy, Health, and Education

This systematic review and meta-analysis of global peer-reviewed neuroimaging findings preliminarily assessed the magnitude of effect sizes (ES) of the influences of family poverty/low socioeconomic status (SES) on children’s neurocognition and whether these were consistently detrimental. The literature search (Web of Science; PUBMED; MEDLINE: PSYCNET; GOOGLE SCHOLAR; SCIENCEDIRECT) included 66 studies from 1988 to 2022; 85% of the studies included were conducted in Western, high-income nations. Bayesian models, corrected by study sizes and variances, revealed ESs were heterogeneous across countries and measurements. Bayesian and standard hypothesis testing indicated high and low SES groups showed similar behavioral performances in neuroimaging-concurrent tasks. Except for Magnetic Resonance Imaging studies, ESs were small-to-intermediate with modest reliability. The strongest ESs were found for attention, mathematical performance, language, and cortical volume, followed by intermediate ESs for reading and socioemotional processes. Differentials in resting activity and connectivity, working memory, and executive functions yielded small effects. A bibliometric analysis showed a significant proportion of the literature attributed neurocognitive deficits to low SES, despite overlooking the under-representativity of non-Western and low-income countries, potential influences of racial/ethnic differences, and measurement sensitivity/specificity discrepancies. To reach United Nations Sustainable Development Goals, policies and interventions should consider regional, structural, or environmental ecologies beyond the individual, critically probing implicit deficit attributions.

Bridge Damage Detection Using Ambient Loads by Bayesian Hypothesis Testing for a Parametric Subspace of an Autoregressive Model

This study explores a change detection method in modal properties to automate and generalize in-service damage detection for vibration-based structural health monitoring of bridges. The noisy conditions caused by ambient loading pose difficulty for in-service damage detection because the load-induced noise often masks the difference in the modal properties. The proposed method directly converts measured time series into a simplified anomaly indicator robust against load-induced noise. This study adopts a vector autoregressive model to represent the vibration of bridges. Bayesian inference produces a posterior probability distribution function of the model parameters. Principal component analysis extracts a subspace comparable to the modal properties in the model parameters. Bayesian hypothesis testing quantifies anomalies in the extracted subspace. The feasibility of the proposed method is assessed with vibration data from field experiments conducted on an actual steel truss bridge. The field experiment includes damage severing the truss members. The modal frequencies and mode shapes estimated from the principal component analysis correspond well to earlier reported results. The proposed damage detection method successfully indicated all damage considered in the experiment.

Bayesian tests for random mating in polyploids.

Hardy-Weinberg proportions (HWP) are often explored to evaluate the assumption of random mating. However, in autopolyploids, organisms with more than two sets of homologous chromosomes, HWP and random mating are different hypotheses that require different statistical testing approaches. Currently, the only available methods to test for random mating in autopolyploids (i) heavily rely on asymptotic approximations and (ii) assume genotypes are known, ignoring genotype uncertainty. Furthermore, these approaches are all frequentist, and so do not carry the benefits of Bayesian analysis, including ease of interpretability, incorporation of prior information, and consistency under the null. Here, we present Bayesian approaches to test for random mating, bringing the benefits of Bayesian analysis to this problem. Our Bayesian methods also (i) do not rely on asymptotic approximations, being appropriate for small sample sizes, and (ii) optionally account for genotype uncertainty via genotype likelihoods. We validate our methods in simulations and demonstrate on two real datasets how testing for random mating is more useful for detecting genotyping errors than testing for HWP (in a natural population) and testing for Mendelian segregation (in an experimental S1 population). Our methods are implemented in Version 2.0.2 of the hwep R package on the Comprehensive R Archive Network https://cran.r-project.org/package=hwep.

Gamma-Aminobutyric Acid and Glutamate Concentrations in the Striatum and Anterior Cingulate Cortex Not Found to Be Associated with Cognitive Flexibility.

Behavioral flexibility and goal-directed behavior heavily depend on fronto-striatal networks. Within these circuits, gamma-aminobutyric acid (GABA) and glutamate play an important role in (motor) response inhibition, but it has remained largely unclear whether they are also relevant for cognitive inhibition. We hence investigated the functional role of these transmitters for cognitive inhibition during cognitive flexibility. Healthy young adults performed two paradigms assessing different aspects of cognitive flexibility. Magnetic resonance spectroscopy (MRS) was used to quantify GABA+ and total glutamate/glutamine (Glx) levels in the striatum and anterior cingulate cortex (ACC) referenced to N-acetylaspartate (NAA). We observed typical task switching and backward inhibition effects, but striatal and ACC concentrations of GABA+/NAA and Glx/NAA were not associated with cognitive flexibility in a functionally relevant manner. The assumption of null effects was underpinned by Bayesian testing. These findings suggest that behavioral and cognitive inhibition are functionally distinct faculties, that depend on (at least partly) different brain structures and neurotransmitter systems. While previous studies consistently demonstrated that motor response inhibition is modulated by ACC and striatal GABA levels, our results suggest that the functionally distinct cognitive inhibition required for successful switching is not, or at least to a much lesser degree, modulated by these factors.

Classical models may be a better explanation of the Jiuzhang 1.0 Gaussian Boson Sampler than its targeted squeezed light model

Recently, Zhong et al. [1, 2] performed landmark Gaussian boson sampling experiments with up to 144 modes using threshold detectors. The authors claim to have achieved quantum computational advantage with the implementation of these experiments, named Jiuzhang 1.0 and Jiuzhang 2.0. Their experimental results are validated against several classical hypotheses and adversaries using tests such as the comparison of statistical correlations between modes, Bayesian hypothesis testing and the Heavy Output Generation (HOG) test. In this work we propose an alternative classical hypothesis for the validation of these experiments. We use the probability distribution of mixtures of coherent states sent into a lossy interferometer; these input mixed states, which we term squashed states, have vacuum fluctuations in one quadrature and excess fluctuations in the other. We find that for configurations in the high photon number density regime, the comparison of statistical correlations does not tell apart the ground truth of the experiment (two-mode squeezed states sent into an interferometer) from our alternative hypothesis. On the other hand, the Bayesian test indicates that, for all configurations excepting Jiuzhang 1.0, the ground truth is a more likely explanation of the experimental data than our alternative hypothesis. A similar result is obtained for the HOG test: for all configurations of Jiuzhang 2.0, the test indicates that the experimental samples have higher ground truth probability than the samples obtained from our alternative distribution; for Jiuzhang 1.0 the test is inconclusive. Our results provide a new hypothesis that should be considered in the validation of future GBS experiments, and shed light into the need to identify proper metrics to verify quantum advantage in the context of threshold GBS. Additionally, they indicate that a classical explanation of the Jiuzhang 1.0 experiment, lacking any quantum features, has not been ruled out.

Sequential Bayesian assurance tests for degradation data

Measuring degradation of items in a reliability context may be all that is feasible. There may simply be neither enough time nor resources to produce a sufficient number of item failures to characterize the underlying time-until-failure distribution. In such contexts, degradation data-based assurance testing can be tuned to strike a compromise between consumer and producer risk when deciding whether to accept or reject a product. A one-stage assurance test counts the number of items in a sample exceeding a fixed degradation threshold at a fixed time and uses this count to make the decision: accept or reject. A general Bayesian framework for extending assurance testing from one-stage to a multi-stage or sequential setting is presented. Our multi-stage assurance tests are shown to compare favorably to their one-stage counterparts by possessing a lower expected time requirement at given sample size and risk constraints. Examples of the methods based on a printhead application are provided and are reproducible with the supplemental R code.

Inferring the Direction of Introgression Using Genomic Sequence Data.

Genomic data are informative about the history of species divergence and interspecific gene flow, including the direction, timing, and strength of gene flow. However, gene flow in opposite directions generates similar patterns in multilocus sequence data, such as reduced sequence divergence between the hybridizing species. As a result, inference of the direction of gene flow is challenging. Here, we investigate the information about the direction of gene flow present in genomic sequence data using likelihood-based methods under the multispecies-coalescent-with-introgression model. We analyze the case of two species, and use simulation to examine cases with three or four species. We find that it is easier to infer gene flow from a small population to a large one than in the opposite direction, and easier to infer inflow (gene flow from outgroup species to an ingroup species) than outflow (gene flow from an ingroup species to an outgroup species). It is also easier to infer gene flow if there is a longer time of separate evolution between the initial divergence and subsequent introgression. When introgression is assumed to occur in the wrong direction, the time of introgression tends to be correctly estimated and the Bayesian test of gene flow is often significant, while estimates of introgression probability can be even greater than the true probability. We analyze genomic sequences from Heliconius butterflies to demonstrate that typical genomic datasets are informative about the direction of interspecific gene flow, as well as its timing and strength.

Handling Imbalanced Classification Problems With Support Vector Machines via Evolutionary Bilevel Optimization.

Support vector machines (SVMs) are popular learning algorithms to deal with binary classification problems. They traditionally assume equal misclassification costs for each class; however, real-world problems may have an uneven class distribution. This article introduces EBCS-SVM: evolutionary bilevel cost-sensitive SVMs. EBCS-SVM handles imbalanced classification problems by simultaneously learning the support vectors and optimizing the SVM hyperparameters, which comprise the kernel parameter and misclassification costs. The resulting optimization problem is a bilevel problem, where the lower level determines the support vectors and the upper level the hyperparameters. This optimization problem is solved using an evolutionary algorithm (EA) at the upper level and sequential minimal optimization (SMO) at the lower level. These two methods work in a nested fashion, that is, the optimal support vectors help guide the search of the hyperparameters, and the lower level is initialized based on previous successful solutions. The proposed method is assessed using 70 datasets of imbalanced classification and compared with several state-of-the-art methods. The experimental results, supported by a Bayesian test, provided evidence of the effectiveness of EBCS-SVM when working with highly imbalanced datasets.

Comparing the prevalence of statistical reporting inconsistencies in COVID-19 preprints and matched controls: a registered report.

The COVID-19 outbreak has led to an exponential increase of publications and preprints about the virus, its causes, consequences, and possible cures. COVID-19 research has been conducted under high time pressure and has been subject to financial and societal interests. Doing research under such pressure may influence the scrutiny with which researchers perform and write up their studies. Either researchers become more diligent, because of the high-stakes nature of the research, or the time pressure may lead to cutting corners and lower quality output. In this study, we conducted a natural experiment to compare the prevalence of incorrectly reported statistics in a stratified random sample of COVID-19 preprints and a matched sample of non-COVID-19 preprints. Our results show that the overall prevalence of incorrectly reported statistics is 9-10%, but frequentist as well as Bayesian hypothesis tests show no difference in the number of statistical inconsistencies between COVID-19 and non-COVID-19 preprints. In conclusion, the literature suggests that COVID-19 research may on average have more methodological problems than non-COVID-19 research, but our results show that there is no difference in the statistical reporting quality.

Evaluating predictors' relative importance using Bayes factors in regression models.

This study presents a Bayesian inference approach to evaluate the relative importance of predictors in regression models. Depending on the interpretation of importance, a number of indices are introduced, such as the standardized regression coefficient, the average squared semipartial correlation, and the dominance analysis measure. Researchers' theories about relative importance are represented by order constrained hypotheses. Support for or against the hypothesis is quantified by the Bayes factor, which can be computed from the prior and posterior distributions of the importance index. As the distributions of the indices are often unknown, we specify prior and posterior distributions for the covariance matrix of all variables in the regression model. The prior and posterior distributions of each importance index can be obtained from the prior and posterior samples of the covariance matrix. Simulation studies are conducted to show different inferences resulting from various importance indices and to investigate the performance of the proposed Bayesian testing approach. The procedure of evaluating relative importance using Bayes factors is illustrated using two real data examples. (PsycInfo Database Record (c) 2021 APA, all rights reserved).