Search

BACKGROUND: While some evidence has potentially linked climate change to carcinogenic factors, the long-term effect of climate change on liver cancer risk largely remains unclear. OBJECTIVES: Our objective is to evaluate the long-term relationship between temperature increase and liver cancer incidence in Australia. METHODS: We mapped the spatial distribution of liver cancer incidence from 2001 to 2019 in Australia. A Bayesian spatial conditional autoregressive (CAR) model was used to estimate the relationships between the increase in temperature at different lags and liver cancer incidence in Australia, after controlling for chronic hepatitis B prevalence, chronic hepatitis C prevalence, and the Index of Relative Socio-economic Disadvantage. Spatial random effects obtained from the Bayesian CAR model were also mapped. RESULTS: The research showed that the distribution of liver cancer in Australia is spatially clustered, most areas in Northern Territory and Northern Queensland have higher incidence and relative risk. The increase in temperature at the lag of 30 years was found to correlate with the increase in liver cancer incidence in Australia, with a posterior mean of 30.57 [95% Bayesian credible interval (CrI): 0.17, 58.88] for the univariate model and 29.50 (95% CrI: 1.27, 58.95) after controlling for confounders, respectively. The results were not highly credible for other lags. DISCUSSION: Our Bayesian spatial analysis suggested a potential relationship between temperature increase and liver cancer. To our knowledge, this research marks the first attempt to assess the long-term effect of global warming on liver cancer. If the relationship is confirmed by other studies, these findings may inform the development of prevention and mitigation strategies based on climate change projections. https://doi.org/10.1289/EHP14574, Introduction Liver cancer is a disease with a heavy burden on patients. Global Burden of Disease Study 2019 (GBD 2019) reported that there were 534,000 new cases and 485,000 deaths [...]

Structural reliability theories are used in the calibration of load and resistance factor design (LRFD) in the Canadian Highway Bridge Design Code (CHBDC). The LRFD approach contains certain assumptions about uncertainties in the load and capacity estimation that prevent it from fully exploiting the information gathered during visual inspections. This paper presents a reliability-based framework for analyzing the visual inspection data obtained according to the Ontario structure inspection manual (OSIM). Existing deterioration models are adapted. The Bayesian interference is utilized to estimate the updated structural properties according to the prior information from the bridge maintenance and deterioration models and the new information collected from visual inspections. The criteria set by the CHBDC are used to analyze components and systems reliability. The value of the proposed framework for bridge evaluation and optimizing maintenance is demonstrated through the full implementation of a case-study bridge in the Canadian Province of NB. Key words: reliability analysis, bridge engineering, visual inspection, Bayesian interference, 1. Introduction Bridges deteriorate over time as a result of exposure to severe loading and environmental conditions. According to a survey by Statistics Canada in 2018, more than 10% of [...]

Quantifying the severity of neurological disorders, such as attention-deficit/hyperactivity disorder (ADHD), is encouraged in practice. To explore the methodological properties of data-driven classification, this study aimed to develop computerized adaptive testing (CAT) based on Bayesian networks to classify ADHD symptom severity for individual subjects efficiently. The Current Symptoms Scale was used to build an ADHD symptom severity classifier in a sample of 892 college students, and the CAT making use of information gains from conditional probabilities of the Bayesian network was evaluated through a simulation study. The results showed that distributing 8 out of 18 items on average to an individual subject would result in 85% classification accuracy, supporting the adaption of surveying ADHD to digital platforms demanding much-shortened scales. The severity classifier based on Bayesian networks can assist diagnostic tasks via a data-driven mechanism. Further, integrating CAT into the classifier can shorten the scale while maintain acceptable measurement quality. The ADHD symptoms were used in the paper; however, the proposed method applies to other psychological symptoms as well. The impact and limitations of the CAT for ADHD were discussed, and several suggestions for future research were provided., Author(s): Zhehan Jiang [sup.1] , Wenchao Ma [sup.2] , Kate Flory [sup.3] , Dan Zhang [sup.4] , Wenjing Zhou [sup.1] , Dexin Shi [sup.3] , Xiaochen Hua [sup.5] , Ren [...]

We investigate the application of ensemble transform approaches to Bayesian inference of logistic regression problems. Our approach relies on appropriate extensions of the popular ensemble Kalman filter and the feedback particle filter to the cross entropy loss function and is based on a well-established homotopy approach to Bayesian inference. The arising finite particle evolution equations as well as their mean-field limits are affine-invariant. Furthermore, the proposed methods can be implemented in a gradient-free manner in case of nonlinear logistic regression and the data can be randomly subsampled similar to mini-batching of stochastic gradient descent. We also propose a closely related SDE-based sampling method which again is affine-invariant and can easily be made gradient-free. Numerical examples demonstrate the appropriateness of the proposed methodologies., Author(s): Jakiw Pidstrigach [sup.1] , Sebastian Reich [sup.1] Author Affiliations: (1) grid.11348.3f, 0000 0001 0942 1117, Institut für Mathematik, Universität Potsdam, , Karl-Liebknecht-Str. 24/25, 14476, Potsdam, Germany Introduction Statistical inference [...]

Wilcox proposed an argument against imprecise probabilities and for the principle of indifference based on a thought experiment where he argues that it is very intuitive to feel that one's confidence in drawing a ball of a given colour out of an unknown urn should decrease while the number of potential colours in the urn increases. In my response to him, I argue that one's intuitions may be unreliable because it is very hard to truly feel completely ignorant in such a situation. I further argue that Wilcox must also account for the conflicting intuition that it is absurd to have to feel completely convinced that a specific claim about reality is true in the absence of any evidence in order to avoid being irrational. It is dubious that this intuition is considerably less universal and strongly-held than Wilcox's own intuition. Finally, I point out that even if Wilcox's intuition were to be universally shared among members of our biological species, it is far from being clear that someone refusing to let that intuition dictate his or her beliefs would be irrational. For all these reasons, I believe that Wilcox was not successful in proving that philosophers and scientists representing uncertainty through imprecise probabilities are violating the principles of rationality., Author(s): Marc Fischer [sup.1] Author Affiliations: (1) grid.25697.3f, 0000 0001 2172 4233, Mines de Saint-Étienne, CNRS, UMR 5307 LGF, Centre SPIN, F-42023 Saint-Étienne, Université de Lyon, , Lyon, France Introduction [...]

Telemetry and mark-recapture provide movement information, but each approach comes with tradeoffs, potentially producing conflicting understandings of fish movement patterns. Using a Bayesian framework that allows exchanging priors from either method may help assess these inconsistencies. We evaluated channel catfish Ictalurus punctatus movements in the Red River of the North and Lake Winnipeg system, which impacts harvest management across different jurisdictions and affects different ecosystems (e.g., lotic and lentic). Channel catfish were tagged with T-bar tags or acoustic transmitters. The resulting movement data were modeled using a Bayesian multi-state Cormack-Jolly-Seber model to estimate survival, movement, and recapture probabilities. Model estimates with uninformative priors showed a greater tendency of downstream movement from the Red River into Lake Winnipeg for the T-bar tags. In contrast, the telemetry method showed fish predominantly stay in the river. However, exchanging increasingly stronger prior information from the alternative method's model revealed that telemetry movement estimates were less sensitive than the T-bar model to informative priors. Using priors from both methods provided a transparent means to assessing tagging approach tradeoffs quantitatively. Key words: Bayesian, multi-state Cormack-Jolly-Seber model, informative priors, telemetry, mark-recapture Si les techniques de telemetrie et de marquage-recapture fournissent de l'information sur les deplacements, des compromis sont associes a chacune de ces approches, qui peuvent mener a des interpretations contradictoires des motifs de deplacements des poissons. L'utilisation d'un cadre bayesien qui permet l'echange de renseignements prealables de l'une a l'autre de ces methodes peut aider a evaluer ces incoherences. Nous avons evalue les deplacements de barbues de riviere Ictalurus punctatus dans le reseau de la riviere Rouge et du lac Winnipeg, qui ont une incidence sur la gestion des prises dans differents territoires et sur differents ecosystemes (p. ex. lotiques et lentiques). Des barbues de riviere ont ete dotees d'etiquettes en T ou d'emetteurs acoustiques. Les donnees de deplacement resultantes ont ete modelisees en utilisant un modele bayesien multi-etats de Cormack-Jolly-Seber afin d'estimer la survie, les deplacements et les probabilites de recapture. Les estimations du modele integrant des renseignements prealables non informatifs montrent une plus forte tendance de deplacements vers l'aval de la riviere Rouge vers le lac Winnipeg pour les etiquettes en T. La telemetrie montre en revanche que les poissons restent de maniere preponderante dans la riviere. L'echange de renseignements prealables de plus en plus forts entre les modeles des deux methodes revele que les estimations des deplacements par telemetrie sont moins sensibles aux renseignements prealables informatifs que celles du modele pour les etiquettes en T. L'utilisation de renseignements prealables issus des deux methodes offre une approche transparente pour evaluer de maniere quantitative les compromis associes aux differentes methodes de marquage. [Traduit par la Redaction] Mots-cles : bayesien, modele mutli-etats de Cormack-Jolly-Seber, renseignements prealables informatifs, telemetrie, marquage-recapture, 1. Introduction Incorporating the spatial dynamics of fisheries into decision-making remains a challenge for fisheries management as management perspectives transition from local to ecosystem scales (Berkeley et al. 2004; Ciannelli [...]

In Georgia, children in high-risk counties are at increased risk for lead exposure. Those children and others in high-risk groups, such as families receiving Medicaid and Peach Care for Kids (i.e., health coverage for children in low-income families), are screened for blood lead levels (BLLs). Such screening, however, might not include all children at high risk for having BLLs above the reference levels ([greater than or equal to] 5 [micro]g/dL) in the state. In our study, Bayesian methods were used to estimate the predictive density of the number of children, Introduction Lead exposure can seriously affect the health of children (World Health Organization, 2022). High levels of lead exposure can harm the brain and central nervous system of children. High [...]

Forest resource assessments based on multi-source and multi-temporal data have become more common. Therefore, enhancing the prediction capabilities of forestry dynamics by efficiently pooling and analyzing time-series and spatial sequential data is now more pivotal. Bayesian filtering and smoothing provide a well-defined formalism for the fusion or assimilation of various data. We ascertained how often the generic, standardized Bayesian framework is used in the scientific literature and whether such an approach is beneficial for forestry applications. A review of the literature showed that the use of Bayesian methods appears to be less common in forestry than in other disciplines, particularly remote sensing. Specifically, time-series analyses were found to favor ad hoc methods. Our review did not reveal strong numeric evidence for better performance by the various Bayesian approaches, but this result may be partly due to the challenge in comparing a variety of methods for different prediction tasks. We identified methodological challenges related to assimilating predictions of forest development; in particular, combining modelled growth with disturbances due to both forest operations and natural phenomena. Nevertheless, the Bayesian frameworks provide possibilities to efficiently combine and update prior and posterior predictive distributions and derive related uncertainty measures that appear under-utilized in forestry. Key words: forest inventory, hierarchical Bayes model, Kalman filter, Markov chain Monte Carlo (MCMC), credible interval. L'evaluation des ressources forestieres fondee sur des donnees de nombreuses sources et multitemporelles est devenue plus frequente. Par consequent, l'amelioration des capacites de previsions de la dynamique forestiere au moyen du regroupement et de l'analyse efficaces des donnees sequentielles spatiales et de series chronologiques est maintenant plus cruciale. Le filtrage et le nivellement bayesiens fournissent un formalisme bien defini pour la fusion ou l'assimilation des diverses donnees. Nous avons determine la frequence a laquelle le cadre bayesien generique et normalise est utilise dans la litterature scientifique et nous avons evalue si une telle approche est avantageuse pour les applications forestieres. Une revue de la litterature a montre que l'utilisation des methodes bayesiennes semble etre moins frequente en foresterie que dans d'autres disciplines, particulierement la teledetection. On a constate plus particulierement que les analyses des series chronologiques favorisaient les methodes ad hoc. Notre revue n'a pas revele de preuves numeriques importantes pour une meilleure performance par diverses approches bayesiennes, mais ce resultat peut etre du en partie a la difficulte de comparer un eventail de methodes pour differentes taches de previsions. Nous avons identifie des defis methodologiques relies a l'assimilation des previsions du developpement forestier; plus particulierement, en combinant la croissance modelisee aux perturbations dues aux operations forestieres et aux phenomenes naturels. Neanmoins, les cadres bayesiens offrent des possibilites de combiner efficacement et de faire une mise a jour avant et apres les distributions predictives et de calculer les mesures connexes de l'incertitude qui semblent sous-utilisees en foresterie. [Traduit par la Redaction] Mots-cles : inventaire forestier, modele hierarchique bayesien, filtre de Kalman, Monte Carlo par chaine de Markov (MCCM), intervalle de credibilite., 1. Introduction 1.1. Motivation and objectives Systematic forest inventories have been carried out for over a century (Kangas et al. 2018a) and have utilized remote sensing and other digital map [...]

Acoustic telemetry systems technology is useful for studying fish movement patterns and habitat use; however, the data generated from omnidirectional acoustic receivers are prone to large observation errors because the tagged animal can be anywhere in the detection range of the receiver. In this study, we used the Bayesian state space modeling (SSM) approach and different smoothing methods including kernel smoothing and cross-validated local polynomial regression to reconstruct fish movement paths of walleye (Sander vitreus) using data obtained from a telemetry receiver grid in Lake Winnipeg, Manitoba, Canada. Using the SSM approach, we obtained more realistic movement paths, compared with the smoothing methods. In addition, we have highlighted the advantages of the SSM approach to estimate undetected movement paths, over simple smoothing techniques, by comparing ecological metrics such as path length and tortuosity between different reconstruction approaches. Reconstructed paths could be useful in making effective fishery management decisions on Lake Winnipeg in the future by providing information on how walleye move and distribute in Lake Winnipeg over space and time. La technologie des systemes de telemetrie acoustique est utile pour l'etude des habitudes de deplacement des poissons et de l'utilisation de l'habitat par ces derniers. Les donnees produites par les recepteurs acoustiques omnidirectionnels se pretent toutefois a d'importantes erreurs d'observation, puisque l'animal etiquete peut se trouver n'importe ou a l'interieur du rayon de detection du recepteur. Nous avons utilise l'approche bayesienne de modelisation d'espace d'etats (MEE) et differentes methodes de lissage, dont le lissage par noyau et la regression locale polynomiale jumelee a la validation croisee, pour reconstituer les trajectoires de deplacement de dores jaunes (Sander vitreus) en utilisant des donnees obtenues d'un reseau de recepteurs telemetriques dans le lac Winnipeg, Manitoba, Canada. En utilisant l'approche de MEE, nous obtenons des trajectoires de deplacement plus realistes que ce que donnent les methodes de lissage. Nous faisons en outre ressortir les avantages de l'approche de MEE pour l'estimation de trajectoires de deplacement non detectees comparativement aux techniques de lissage simples, en comparant des parametres ecologiques comme la longueur et la tortuosite des trajectoires pour differentes approches de reconstitution. Les trajectoires reconstituees pourraient etre utiles a une prise de decisions de gestion efficace pour les peches dans le lac Winnipeg en fournissant de l'information sur les deplacements et la repartition dans l'espace et le temps des dores jaunes dans ce lac. [Traduit par la Redaction], Introduction Fish can be highly mobile in their search for suitable foraging and spawning habitats (Cooke et al. 2016). Fish movement paths reveal habitat use, migration patterns, survival, and behavioural [...]

Coronavirus disease (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARSCoV-2) (1), was detected in the United States in January 2020 (2). Researchers documented deaths in the United States [...]

This paper discusses a Bayesian approach to analyzing cost efficiency of Distribution System Operators when model specification and variable selection are difficult to determine. Bayesian model selection and inference pooling techniques are adopted in a stochastic frontier analysis to mitigate the problem of model uncertainty. Adequacy of a given specification is judged by its posterior probability, which makes the benchmarking process not only more transparent but also much more objective. The proposed methodology is applied to one of Polish Distribution System Operators. We find that variable selection plays an important role and models, which are the best at describing the data, are rather parsimonious. They rely on just a few variables determining the observed cost. However, these models also show relatively high average efficiency scores among analyzed objects. Keywords: Cost efficiency, Electricity distribution, Bayesian inference, Stochastic frontier analysis, 1. INTRODUCTION Electricity distribution is a key sector of every economy. In most countries, distribution system operators of electricity networks (DSOs hereafter) are natural, usually territorial, monopolies. They are either [...]

We study the optimal pricing problem of a monopolistic firm facing customers with limited attention and capability to process information about the value (quality) of a single offered product. We model customer choice based on the theory of rational inattention in the economics literature, which enables us to capture not only the impact of true quality and price, but also the intricate effects of customer's prior beliefs and cost of information acquisition and processing. We formulate the firm's price optimization problem assuming that the firm can also use the price to signal the quality of the product to customers. To delineate the economic incentives of the firm, we first characterize the pricing and revenue implications of customer's limited attention without signaling, and then use these results to explore perfect Bayesian equilibria of the strategic pricing signaling game. As an extension, we consider heterogeneous customers with different information costs as well as prior beliefs. We discuss the managerial implications of our key findings and prescribe insights regarding information provision and product positioning. History: Accepted by Gad Allon, operations management. Keywords: pricing * choice behavior * rational inattention * information acquisition * signaling game, 1. Introduction We are living in an era characterized by abundance of information. Firms are in a race to make use of the 'big data' available on their customers and [...]

A total of 29 U.S. states and the District of Columbia have in place mandatory Renewable Portfolio Standards (RPS) which require that a minimum amount of energy come from renewable resources. We investigate the role of hydropower vis-a-vis other renewables under RPS. Using a Bayesian multilevel model, we find that hydropower plants subject to RPS are more likely to plan upgrades. These planned upgrades appear to be a substitute for solar and wind rather than complementary reserve generation. Keywords: Renewable portfolio standards, Hydropower, Bayesian methods, Multilevel models, Planned investment, 1. INTRODUCTION We investigate the effect of Renewable Portfolio Standards (RPS) on planned investments in hydropower upgrades. We also ask whether hydropower is a substitute or complement to other renewables. [...]

Abstract. Technology implementation at the individual level within an organization, after the organization has adopted the technology, has been an ongoing challenge in every field. In this study, we develop [...]

We develop a Bayesian methodology for systemic risk assessment in financial networks such as the interbank market. Nodes represent participants in the network, and weighted directed edges represent liabilities. Often, for every participant, only the total liabilities and total assets within this network are observable. However, systemic risk assessment needs the individual liabilities. We propose a model for the individual liabilities, which, following a Bayesian approach, we then condition on the observed total liabilities and assets and, potentially, on certain observed individual liabilities. We construct a Gibbs sampler to generate samples from this conditional distribution. These samples can be used in stress testing, giving probabilities for the outcomes of interest. As one application we derive default probabilities of individual banks and discuss their sensitivity with respect to prior information included to model the network. An R package implementing the methodology is provided. History: Accepted by Noah Gans, stochastic models and simulation. Supplemental Material: The online supporting document is available at https://doi.org/10.1287/ mnsc.2016.2546. Keywords: financial network * unknown interbank liabilities * systemic risk * Bayes * MCMC * Gibbs sampler * power law, 1. Introduction Assessing systemic risk in financial systems is a key concern for regulators and policy makers. We think of systemic risk as the risk that some external or economic [...]

Link recommendation, which suggests links to connect currently unlinked users, is a key functionality offered by major online social networks. Salient examples of link recommendation include 'People You May Know' on Facebook and LinkedIn as well as 'You May Know' on Google+. The main stakeholders of an online social network include users (e.g., Facebook users) who use the network to socialize with other users and an operator (e.g., Facebook Inc.) that establishes and operates the network for its own benefit (e.g., revenue). Existing link recommendation methods recommend links that are likely to be established by users but overlook the benefit a recommended link could bring to an operator. To address this gap, we define the utility of recommending a link and formulate a new research problem--the utility-based link recommendation problem. We then propose a novel utility-based link recommendation method that recommends links based on the value, cost, and linkage likelihood of a link, in contrast to existing link recommendation methods that focus solely on linkage likelihood. Specifically, our method models the dependency relationship between the value, cost, linkage likelihood, and utility-based link recommendation decision using a Bayesian network; predicts the probability of recommending a link with the Bayesian network; and recommends links with the highest probabilities. Using data obtained from a major U.S. online social network, we demonstrate significant performance improvement achieved by our method compared with prevalent link recommendation methods from representative prior research. History: Accepted by Anandhi Bharadwaj, information systems. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2016.2446. Keywords: utility-based link recommendation * link prediction * Bayesian network learning * continuous latent factor * online social network * machine learning * network formation, 1. Introduction Online social networks such as Facebook, LinkedIn, and Google+ have gained unprecedented numbers of users in a short time, attracting massive attention from both industry and academia to [...]

The switching multiple disorder problem seeks to determine an ordered infinite sequence of times of alarms which are as close as possible to the unknown times of disorders, or change-points, [...]

This study examines if racial identity and gender were related to juvenile officers' perceptions of themselves as advocates for youth and to determine whether bureaucratic representation, such as minority staff, supports rehabilitation for youth. This study also investigates whether incarcerated youth perceive juvenile correctional officers as role models. Using data from the Performance Based Standards project, Bayesian Networks are used to find possible relationships among the different variables involved in this study. The created Bayesian Networks helps to learn more about the relationship between juvenile correctional officers and incarcerated youth. Keywords: Representative bureaucracy, Bayesian Networks, officer's perceptions, juvenile violence, juvenile justice, INTRODUCTION Juvenile correctional staff plays a key role in juvenile institutions by providing security and counseling to confined youth. Officers are responsible for portraying dual roles within juvenile institutions. Therefore, [...]

We investigate the interplay between learning effects and externalities in the problem of competitive investments with uncertain returns. We examine a game theoretic duopoly investment model in which (i) a firm can learn about the profitability of the investment by observing the performance of the first mover and (ii) externalities exist between the investments of two firms. We find a region of a war of attrition between the two firms in which the interplay between externalities and learning gives rise to counterintuitive effects on investment strategies and payoffs. In particular, we find that, contrary to the conventional war of attrition where an increase in benefits for the follower generally delays the first move, an increase in the rate of learning--which tends to benefit the follower--can hasten the first investment. Keywords: games; group decisions; stochastic; decision analysis; sequential; dynamic programming; optimal control History: Received December 11, 2013; accepted January 7, 2015, by James Smith, decision analysis. Published online in Articles in Advance August 18, 2015., 1. Introduction Investment decisions in business, such as the introduction of products in unproven markets or the adoption of unproven technologies, often involve significant uncertainty associated with the market reception [...]

Because information collected in a site investigation is limited, it is not possible to obtain actual values for the mean, standard deviation, and scale of fluctuation for a soil property of interest. The deviation between the estimated values and the actual values is called the statistical uncertainty. There are at least two schools of thought on how to model the statistical uncertainty: frequentist thought and Bayesian thought. The purpose of this paper is to discuss their philosophical difference, to show how to quantify the statistical uncertainty based on these two distinct schools of thought, and to compare their performances. To quantify the statistical uncertainty, the confidence interval will be used for the frequentist school of thought, whereas the posterior probability distribution will be used for the Bayesian school of thought. Examples will be presented to compare the performances of these two schools of thought in terms of their consistencies. The results show that, in general, the Bayesian thought performs better in terms of consistency. In particular, the Markov chain Monte Carlo method is recommended when the amount of information available is very limited. Key words: statistical uncertainty, site investigation, frequentist, Bayesian, reliability. En raison du manque de donnees provenant d'etudes sur le terrain, il est impossible d'obtenir les valeurs reelles de la moyenne, de l'ecart-type et de l'echelle de variation d'un parametre du sol que l'on souhaite etudier. L'ecart entre les valeurs estimees et les valeurs reelles est appele <>. Il existe au moins deux ecoles de pensee en ce qui concerne la maniere de modeliser l'incertitude statistique : l'ecole frequentiste et l'ecole bayesienne. Dans le present article, on examine la difference de philosophie entre ces deux ecoles, on decrit la methode de calcul de l'incertitude statistique utilisee par chacune des deux ecoles et on compare les performances de ces methodes. Dans le cas de l'ecole frequentiste, l'intervalle de confiance servira a calculer l'incertitude statistique, alors que, dans le cas de l'ecole bayesienne, on utilisera la distribution des probabilites posterieures pour effectuer ce meme calcul. On fournit des exemples permettant de comparer les performances des methodes de calcul utilisees par les deux ecoles en ce qui a trait a leurs consistances respectives. Les resultats montrent en general que les methodes de l'ecole bayesienne sont plus performantes en termes de consistance. En particulier, on recommande d'utiliser la methode de Monte-Carlo par chaines de Markov lorsque la quantite de donnees disponibles est tres limitee. [Traduit par la Redaction] Mots-cles: incertitude statistique, etude sur le terrain, frequentiste, ecole bayesienne, fiabilite., Introduction One of the purposes of site investigation is to obtain information on the spatial distribution of geotechnical design parameters. Typically, the spatial distribution is expressed as a trend function [...]

Data to inform fisheries management are often limited in remote or low economic value fisheries. Here, we use a Bayesian hierarchical modeling structure and two alternative migration timing models to estimate the annual escapement of coho salmon (Oncorhynchus kisutch) to the Chignik River (Alaska, USA) in years with little data, borrowing information from data-rich years to inform parameter estimates. Additionally, we examined trends in peak migration timing between 1922 and 2013 and relative to environmental conditions. Our analyses show that annual escapement estimates are prone to substantial errors unless daily escapement is enumerated for at least 7 days after peak migration date. Finally, peak migration date was negatively correlated with the strength of the Pacific Decadal Oscillation in May-August, and increased over time, though the significance of these associations was dependent on the specific form of the migration timing model used. The modeling approach we present here is easily adaptable to similar situations where data from alternative periods of time or spatial locations can be used to objectively inform local parameter estimates of population characteristics. Les donnees servant a eclairer la gestion des peches sont souvent limitees en ce qui concerne les peches eloignees ou de faible valeur economique. Nous utilisons une structure de modelisation hierarchique bayesienne et deux modeles distincts du moment de la migration pour estimer l'echappement annuel de saumons cohos (Oncorhynchus kisutch) dans la riviere Chignik (Alaska, Etats-Unis) durant des annees pour lesquelles peu de donnees sont disponibles, en empruntant de l'information des annees riches en donnees pour eclairer les estimations de parametres. Nous examinons egalement les tendances de la date du maximum de la migration de 1922 a 2013 et par rapport aux conditions environnementales. Nos analyses montrent que les estimations de l'echappement annuel sont assujetties a d'importantes erreurs si l'echappement quotidien n'est pas enumere pour au moins 7 jours apres la date du maximum de la migration. Enfin, la date du maximum de la migration presentait une correlation negative avec la force de l'oscillation decennale du Pacifique de mai a aout et augmente dans le temps, la signification de ces associations dependant toutefois de la forme precise du modele du moment de la migration employe. L'approche de modelisation que nous presentons peut facilement etre adaptee a des situations semblables dans lesquelles des donnees pour d'autres periodes ou d'autres emplacements peuvent etre utilisees pour eclairer objectivement les estimations de parametres locaux de caracteristiques des populations. [Traduit par la Redaction], Introduction Fishery managers have long recognized the importance of maintaining a sufficient stock of spawning adults to ensure the production ofthe next generation of individuals (e.g., Russell 1931; Beverton and [...]

We develop and implement a Bayesian semiparametric model of demand under interproduct competition that enables us to assess the respective contributions of brand-SKU (stock keeping unit) hierarchy and interproduct similarity to explaining and predicting demand. To incorporate brand-SKU hierarchy effects, we use Bayesian hierarchical clustering inherent in a nested Dirichlet process to simultaneously partition brands, and SKUs conditional on brands, into groups of 'similarity clusters.' We examine cluster memberships and postprocess the Markov chain Monte Carlo output to infer cluster properties by accounting for parameter uncertainty. Our proposed approach lends to a spatial competition interpretation in latent attribute space and helps uncover the extent to which competition across SKUs in the latent attribute space is local or global. In a related vein, we discuss the implications of well-defined groups of similar SKUs as subcategory or submarket boundaries in latent attribute space. We empirically test our model using aggregate beer category sales data from a midsize U.S. retail chain. We find that branding hierarchy effects dominate those from product similarity. We find that the model partitions the 15 brands in the data into 4 brand clusters and the 96 SKUs into 25 SKU clusters conditional on brand cluster membership. In estimating a set of models of spatial interproduct competition, we find that SKU competition is more local than global in that only subsets of products compete within groups of comparable products. Finally, we discuss the substantive implications of our results. Keywords: competition; nested Dirchlet process; brand-SKU clustering; product similarity; spatial models History: Received June 29, 2012; accepted March 6, 2014, by Pradeep Chintagunta, marketing. Published online in Articles in Advance February 19, 2015., 1. Introduction Competitive effects are integral to market outcomes (e.g., unit sales), particularly in the consumer packaged goods (CPG) industry where there is abundant brand and stock keeping unit (SKU) [...]

After a geotechnical design has been developed, it is common to monitor performance during construction using the observational method by Peck (published in 1969). The observational method is a process where data are collected and geotechnical models updated, allowing timely decisions to be made with respect to risk and opportunity by asset owners or contractors. The observational method is similar to the mathematical formulation for Bayesian updating of material parameters based on measurements. A proof of concept study has been performed to assess the potential for Bayesian updating to be combined with the observational method to allow timely and accurate decision-making during construction of embankments on soft soils. The method was able to converge to an accurate solution prior to 50% consolidation assuming small measurement errors. It is also demonstrated that confidence in the predicted settlement is relatively low at the prior 'design' stage and rapidly increases with three or four measurements spaced over time during the posterior 'construction' phase. Key words: statistical analysis, Bayesian probability analysis, embankments. Apres avoir realise une etude de conception geotechnique, on mesure habituellement les performances durant la phase de construction a l'aide de la methode observationnelle de Peck (publiee en 1969). Celle-ci consiste en un processus durant lequel des donnees sont recueillies et les modeles geotechniques mis a jour, ce qui permet la prise rapide de decisions par les proprietaires et entrepreneurs. La methode observationnelle est similaire a la formulation mathematique d'adaptation bayesienne des parametres de materiau basee sur des mesures. Une etude de validation de principe a ete effectuee pour evaluer la capacite de l'adaptation bayesienne a etre combinee a la methode observationnelle dans le but permettre la prise rapide et precise de decisions lors de la construction de talus sur sols mous. La methode a ete en mesure de deboucher sur une solution precise avant la consolidation a 50 %, en supposant la presence de petites erreurs de mesure. On a egalement demontre que le degre de confiance accorde a la prediction de tassement est relativement faible lors de l'etape prealable de << conception >> et qu'il augmente rapidement en presence de trois ou quatre mesures espacees dans le temps durant la phase ulterieure de << construction >>. [Traduit par la Redaction] Mots-cles : analyse statistique, analyse bayesienne de probabilite, construction de talus., Introduction After a geotechnical design has been performed, it is common to monitor performance during construction using the observational method (Peck 1969). The observational method is a process where data [...]

V. APPLICATION TO PEOPLE V. PUCKETT On February 21, 2008, John Puckett was found guilty of first-degree murder for the 1972 death of Diana Sylvester. (70) Sylvester was a twenty-two-year-old [...]

Understanding and reducing the incidence of accidental bycatch, particularly for vulnerable species such as sharks, is a major challenge for contemporary fisheries management. Here we establish integrated nested Laplace approximations (INLA) and stochastic partial differential equations (SPDE) as two powerful tools for modelling patterns of bycatch through time and space. These novel, computationally fast approaches are applied to fit zero-inflated hierarchical spatiotemporal models to Greenland shark (Somniosus microcephalus) bycatch data from the Baffin Bay Greenland halibut (Reinhardtius hippoglossoides) gillnet fishery. Results indicate that Greenland shark bycatch is clustered in space and time, varies significantly from year to year, and there are both tractable factors (number of gillnet panels, total Greenland halibut catch) and physical features (bathymetry) leading to the high incidence of Greenland shark bycatch. Bycatch risk could be reduced by limiting access to spatiotemporal hotspots or by establishing a maximum number of panels per haul. Our method explicitly models the spatiotemporal correlation structure inherent in bycatch data at a very reasonable computational cost, such that the forecasting of bycatch patterns and simulating conservation strategies becomes more accessible. Comprendre et minimiser les repercussions relatives aux prises accessoires accidentelles, et plus particulierement celles touchant les especes vulnerables, tel le requin, constituent de nos jours un defi de taille en gestion halieutique. Par la presente recherche, nous soumettons que les approximations integrees et agglomerees de Laplace (INLA) ainsi que les equations differentielles partielles stochastiques (EDPS), representent de puissants outils de modelisation de la repartition des prises accessoires dans l'espace et le temps. Ces nouvelles approches informatiques, efficaces et innovantes sont ici utilisees afin de reproduire des modeles spatiotemporels hierarchiques a inflation zero en les appliquant aux donnees relatives aux prises accessoires du requin du Groenland (Somniosus microcephalus) dans la peche au filet maillant du fletan noir (Reinhardtius hippoglossoides) de la baie de Baffin. Les resultats obtenus demontrent que les prises accessoires du requin du Groenland sont dispersees dans l'espace et le temps et varie de maniere considerable selon les annees. Certains facteurs quantifiables, tels le nombre de panneaux de filet par trait et la quantite de prises totale de fletans noirs, ainsi que certaines caracteristiques physiques, telle la bathymetrie, influent sur la frequence des prises accessoires du requin du Groenland. Consequemment, les probabilites de prises accessoires pourraient etre reduites en limitant l'acces de peche dans les endroits spatiotemporels a haut risque ou encore, en etablissant une quantite maximum de panneaux de filet par trait. Ces methodes modelisent d'une facon nette et precise la structure des correlations spatiotemporelles inherentes aux donnees sur les prises accessoires, et ce, a un cout informatique des plus raisonnables, rendant desormais possible et accessible le calcul des previsions quant au nombre de prises accessoires ainsi que la planification de strategies efficaces en conservation des especes., Introduction High levels of bycatch are a predominant problem in many fisheries around the world and contribute greatly to broader concern about overfishing (Kelleher 2005; Davies et al. 2009). Bycatch [...]

Improved methods are needed to evaluate barriers and traps for control and assessment of invasive sea lamprey (Petromyzon marinus) in the Great Lakes. A Bayesian state-space model provided reach-specific probabilities of movement, including trap capture and dam passage, for 148 acoustic tagged invasive sea lamprey in the lower Cheboygan River, Michigan, a tributary to Lake Huron. Reach-specific movement probabilities were combined to obtain estimates of spatial distribution and abundance needed to evaluate a barrier and trap complex for sea lamprey control and assessment. Of an estimated 21828 29 300 adult sea lampreys in the river, 0%-2%, or 0-514 untagged lampreys, could have passed upstream of the dam, and 46%- 61% were caught in the trap. Although no tagged lampreys passed above the dam (0/148), our sample size was not sufficient to consider the lock and dam a complete barrier to sea lamprey. Results also showed that existing traps are in good locations because 83%-96% of the population was vulnerable to existing traps. However, only 52%-69% of lampreys vulnerable to traps were caught, suggesting that traps can be improved. The approach used in this study was a novel use of Bayesian state- space models that may have broader applications, including evaluation of barriers for other invasive species (e.g., Asian carp (Hypophthalmichthys spp.)) and fish passage structures for other diadromous fishes. Il est necessaire d'ameliorer les methodes d'evaluation des barrieres et pieges utilises pour controler et evaluer la lamproie marine (Petromyzon marinus), une espece envahissante, dans les Grands Lacs. Un modele bayesien d'espace d'etats a produit des probabilites de deplacement, dont le piegeage et le passage de barrage, dans differents biefs pour 148 lamproies marines envahissantes dotees de marques acoustiques dans le cours inferieur de la riviere Cheboygan (Michigan), un affluent du lac Huron. Les probabilites de deplacement propres au bief ont ete combinees pour obtenir les estimations de la repartition spatiale et de l'abondance requises pour evaluer un complexe de barriere et piege pour le controle et l'evaluation des lamproies. Sur un nombre estime de 21828 a 29 300 lamproies adultes dans la riviere, de 0 % a 2 % des lamproies non marquees, soit de 0 a 514 poissons, pourraient etre passees en amont du barrage, et de 46 % a 61 % ont ete prises dans le piege. Bien qu'aucune lamproie marquee ne soit passee en amont du barrage (0/148), la taille de l'echantillon est trop petite pour permettre d'etablir que l'ecluse-barrage constitue une barriere completement etanche au passage des lamproies. Les resultats ont egalement demontre que les pieges existants sont bien situes, puisque de 83 % a 96 % de la population etait vulnerable aux pieges existants. Cependant, de 52 % a 69 % seulement des lamproies vulnerables aux pieges ont ete capturees, indiquant que ces pieges peuvent etre ameliores. L'approche employee constitue une utilisation novatrice des modeles bayesiens d'espace d'etats qui pourrait avoir des applications plus larges, notamment pour l'evaluation des barrieres au passage d'autres especes envahissantes (p. ex. carpe asiatique (Hypophthalmichthys spp.)) et des passes a poissons pour d'autres especes diadromes. [Traduit par la Redaction], Introduction The Laurentian Great Lakes were severely disrupted by the arrival of invasive sea lamprey (Petromyzon marinus) through the Welland Canal in the early 1900s (Smith and Tibbles 1980; Hansen [...]