Improved objective Bayesian estimator for a PLP model hierarchically represented subject to competing risks under minimal repair regime

In this paper, we propose a hierarchical statistical model for a single repairable system subject to several failure modes (competing risks). The paper describes how complex engineered systems may be modelled hierarchically by use of Bayesian methods. It is also assumed that repairs are minimal and each failure mode has a power-law intensity. Our proposed model generalizes another one already presented in the literature and continues the study initiated by us in another published paper. Some properties of the new model are discussed. We conduct statistical inference under an objective Bayesian framework. A simulation study is carried out to investigate the efficiency of the proposed methods. Finally, our methodology is illustrated by two practical situations currently addressed in a project under development arising from a partnership between Petrobras and six research institutes.

The Filter Detection Task for measurement of breathing-related interoception and metacognition

The study of the brain’s processing of sensory inputs from within the body (‘interoception’) has been gaining rapid popularity in neuroscience, where interoceptive disturbances have been postulated to exist across a wide range of chronic physiological and psychological conditions. Here we present a task and analysis procedure to quantify specific dimensions of breathing-related interoception, including interoceptive sensitivity (accuracy), decision bias, metacognitive bias, and metacognitive performance. We describe a task that is tailored to methods for assessing respiratory interoceptive accuracy and metacognition, and pair this with an established hierarchical statistical model of metacognition (HMeta-d) to overcome significant challenges associated with the low trial numbers often present in interoceptive experiments. Two major new developments have been incorporated into this task analysis by pairing: (i) a novel adaptive algorithm to maintain task performance at 70-75% accuracy, and (ii) an extended metacognitive model developed to hierarchically estimate multiple regression parameters linking metacognitive performance to relevant (e.g. clinical) variables. We demonstrate the utility of both developments, using both simulated and empirical data from three separate studies. This methodology represents an important step towards accurately quantifying interoceptive dimensions from a simple experimental procedure that is compatible with the practical constraints in clinical settings. Both the task and analysis code are publicly available.

Correlations of Permeability and Geological Characteristics Based on Mercury Intrusion Data and Hierarchical Statistical Models: A Case Study

The mercury intrusion technique is a crucial in-lab method to investigate the porous medium properties. The potentiality of mercury intrusion data has not been explored significantly in the traditional interpretation. Thus, a hierarchical statistical model that not only captures the quantitative relationship between petrophysical properties but also accounts for different geological members is developed to interpret mercury intrusion data. This multilevel model is established from almost 800 samples with specific geological characteristics. We distinguish the fixed effects and the random effects in this mixed model. The overall connection between the selected petrophysical parameters is described by the fixed effects at a higher level, while variations due to different geological members are accommodated as the random effects at a lower level. The selected petrophysical parameters are observed through hypothesis testing and model selection. In this case study, five petrophysical parameters are selected into the model. Essential visualizations are also provided to assist the interpretations of the probabilistically model. The final model reveals the quantitative relationship between permeability and other petrophysical properties in each member and the order of relative importance for each property. With this studied relationship and advanced model, the geological reservoir simulation can be greatly detailed and accurate in the future.

Topic Modeling of Online Accommodation Reviews via Latent Dirichlet Allocation

There is a lot of attention given to the determinants of guest satisfaction and consumer behavior in the tourism literature. While much extant literature uses a deductive approach for identifying guest satisfaction dimensions, we apply an inductive approach by utilizing large unstructured text data of 104,161 online reviews of Korean accommodation customers to frame which topics of interest guests find important. Using latent Dirichlet allocation, a generative, Bayesian, hierarchical statistical model, we extract and validate topics of interest in the dataset. The results corroborate extant literature in that dimensions, such as location and service quality, are important. However, we extend existing dimensions of importance by more precisely distinguishing aspects of location and service quality. Furthermore, by comparing the characteristics of the accommodations in terms of metropolitan versus rural and the type of accommodation, we reveal differences in topics of importance between different characteristics of the accommodations. Specifically, we find a higher importance for points of competition and points of uniqueness among the accommodation characteristics. This has implications for how managers can improve customer satisfaction and how researchers can more precisely measure customer satisfaction in the hospitality industry.

MetPC: Metabolite Pipeline Consisting of Metabolite Identification and Biomarker Discovery Under the Control of Two-Dimensional FDR

Due to the complex features of metabolomics data, the development of a unified platform, which covers preprocessing steps to data analysis, has been in high demand over the last few decades. Thus, we developed a new bioinformatics tool that includes a few of preprocessing steps and biomarker discovery procedure. For metabolite identification, we considered a hierarchical statistical model coupled with an Expectation–Maximization (EM) algorithm to take care of latent variables. For biomarker metabolite discovery, our procedure controls two-dimensional false discovery rate (fdr2d) when testing for multiple hypotheses simultaneously.

Estimating serotype-specific efficacy of pneumococcal conjugate vaccines using hierarchical models

Pneumococcal conjugate vaccines (PCVs) target 10 or 13 specific serotypes. To evaluate vaccine efficacy for these products, the vaccine-targeted serotypes are typically aggregated into a single group to estimate an overall effect. However, it is often desirable to evaluate variations in effects for different serotypes. These serotype-specific estimates are often based on small numbers, resulting in a high degree of uncertainty and instability in the individual estimates. A better approach is to use a Bayesian hierarchical statistical model, which estimates an overall effectiveness of the vaccine across all vaccine-targeted serotypes but also allows the effect to vary by serotype. We re-analyzed published data from a large randomized controlled trial on the efficacy of PCV13 against non-bacteremic community-acquired pneumonia caused by vaccine-targeted serotype. This model provides a potential framework for obtaining more credible and stable estimates of serotype-specific vaccine efficacy and effectiveness.

African attitudes toward foreign countries: A hierarchical approach

This study uses a hierarchical statistical model to explore what drives African attitudes toward foreign countries, an understudied and underappreciated topic in the fields of African studies, political psychology and international relations. Enabled by increasing access to information, technology and resources, African peoples across the continent are directing their national stages like never before. Yet when it comes to international engagements on the continent, there is little scholarly focus on African sentiment toward foreign countries. The present study finds that the drivers of attitudes toward foreign countries vary, depending on which foreign country is under consideration by the respondent. For China, positive sentiment is associated with individuals who report having assets, belong to the president’s party, view domestic governance positively, are more politically interested, are more trusting, are educated and are more frequent news consumers. While political interest, trust, education, news consumption and positive evaluation of domestic governance also correspond to positive attitudes toward the USA, so does being employed and perceiving the domestic government as corrupt. When it comes to attitudes toward former colonies, trust, education, news consumption, positive evaluations of domestic governance, perceiving the domestic government as corrupt, employment, identification with the president’s party and support for democracy are associated with positive sentiment. For China, the USA and former colonies, negative sentiment is associated with individuals who identify as Muslim.