[pt] Desde fevereiro de 2020 a pandemia gerada pelo novo coronavírus SarsCoV-2, vírus gerador da doença COVID-19, tem causado muitos óbitos, principalmente nos grandes centros urbanos. No Brasil, um dos estados mais afetados foi o Rio de Janeiro que, apesar de todas as ações feitas para mitigar o avanço da COVID-19, chegou em 01 de março de 2021 a uma taxa de mortalidade de 206,9 por cento, que corresponde a aproximadamente 207 óbitos a cada mil habitantes. No entanto, os municípios do RJ foram atingidos de maneira distinta, onde a cidade menos afetada alcançou 9,7 por cento e a mais afetada 331,3 por cento. Estudos prévios da literatura especializada indicam que a principal razão desta discrepância pode ser associada à fatores relacionados a população, renda, educação, saúde, economia, território e ambiente. Portanto, esse trabalho tem como principal objetivo identificar os principais fatores socioeconômicos, sociodemográficos e de acesso a recursos hospitalares que estão associadas a taxa de mortalidade oriunda do Sars-CoV-2 nos noventa e dois municípios do estado do Rio de Janeiro com base no modelo de Regressão de Poisson, no período de 01 de março de 2020 a 01 de março de 2021, contabilizando 12 meses. A partir do modelo escolhido foi possível detectar que dez dos onze fatores analisados influenciam na taxa de mortalidade. Sendo os fatores, Índice de desenvolvimento humano municipal (IDHM), Renda per capita (RDPC), Percentual de pobres (PMPOB), Produto interno bruto (PIB), Taxa de frequência bruta ao superior (T_FBSUPER), percentual de aglomerados subnormais (PER_AGSN), Densidade demográfica, Número de leitos hospitalares do SUS por habitante, Número de leitos hospitalares totais por habitante e Número de respiradores por habitante. Assim, os resultados obtidos com base nesses fatores analisados podem auxiliar na criação de ações mitigadoras mais direcionadas e eficientes, de acordo com as características dos municípios do estado do Rio de Janeiro. / [en] Since February 2020 the pandemic generated by the new coronavirus SarsCoV-2, the virus generating the disease COVID-19, has caused many deaths, mainly in large urban centers. In Brazil, one of the most affected states was Rio de Janeiro, which, despite all the actions taken to mitigate the progress of COVID19, reached on March 1, 2021 a mortality rate of 206.9 percent, which corresponds to approximately 207 deaths per thousand inhabitants. However, the Rio de Janeiro municipalities were affected differently, where the least affected city reached 9.7 percent and the most affected 331.3 percent. Previous studies in the specialized literature indicate that the main reason for this discrepancy may be associated with factors related to population, income, education, health, economy, territory, and environment. Therefore, this work has as main objective to identify the main socioeconomic, socio-demographic factors and access to hospital resources that are associated with the mortality rate from Sars-CoV-2 in the ninety-two municipalities in the state of Rio de Janeiro based on the Poisson Regression model, in the period from March 01, 2020 to March 01, 2021, accounting for 12 months. From the model chosen it was possible to detect those ten of the eleven factors analyzed influence the mortality rate. The factors being, municipal human development index (IDHM), per capita income (RDPC), percentage of poor (PMPOB), gross domestic product (GDP), gross attendance rate to higher (T_FBSUPER), percentage of subnormal settlements (PER_AGSN), demographic density, number of SUS hospital beds per inhabitant, number of total hospital beds per inhabitant and number of respirators per inhabitant. Thus, the results obtained based on these analyzed factors can help in the creation of more targeted and efficient mitigating actions, according to the characteristics of the municipalities in the state of Rio de Janeiro.

In epidemiology it is often useful to monitor disease occurrences prospectively to determine the location and time when clusters of disease are forming. This aids in the prevention of illness and injury of the public and is the reason spatio-temporal disease surveillance methods are implemented. Care must be taken in the design and implementation of these types of surveillance methods so that the methods provide accurate information on the development of clusters. Here two spatio-temporal methods for prospective disease surveillance are considered. These include the local Knox monitoring method and a new wavelet-based prospective monitoring method. The local Knox surveillance method uses a cumulative sum (CUSUM) control chart for monitoring the local Knox statistic, which tests for space-time clustering each time there is an incoming observation. The detection of clusters of events occurring close together both temporally and spatially is important in finding outbreaks of disease within a specified geographic region. The local Knox surveillance method is based on the Knox statistic, which is often used in epidemiology to test for space-time clustering retrospectively. In this method, a local Knox statistic is developed for use with the CUSUM chart for prospective monitoring so that epidemics can be detected more quickly. The design of the CUSUM chart used in this method is considered by determining the in-control average run length (ARL) performance for different space and time closeness thresholds as well as for different control limit values. The effect of nonuniform population density and region shape on the in-control ARL is explained and some issues that should be considered when implementing this method are also discussed. In the wavelet-based prospective monitoring method, a surface of incidence counts is modeled over time in the geographical region of interest. This surface is modeled using Poisson regression where the regressors are wavelet functions from the Haar wavelet basis. The surface is estimated each time new incidence data is obtained using both past and current observations, weighing current observations more heavily. The flexibility of this method allows for the detection of changes in the incidence surface, increases in the overall mean incidence count, and clusters of disease occurrences within individual areas of the region, through the use of control charts. This method is also able to incorporate information on population size and other covariates as they change in the geographical region over time. The control charts developed for use in this method are evaluated based on their in-control and out-of-control ARL performance and recommendations on the most appropriate control chart to use for different monitoring scenarios is provided. / Ph. D.

Modeling and discovering knowledge from multi-relational and multivariate discrete data is a crucial task that arises in many research and application domains, e.g. text mining, intelligence analysis, epidemiology, social science, etc. In this dissertation, we study and address three problems involving the modeling of multi-relational discrete data and multivariate multi-response count data, viz. (1) discovering surprising patterns from multi-relational data, (2) constructing a generative model for multivariate categorical data, and (3) simultaneously modeling multivariate multi-response count data and estimating covariance structures between multiple responses. To discover surprising multi-relational patterns, we first study the ``where do I start?'' problem originating from intelligence analysis. By studying nine methods with origins in association analysis, graph metrics, and probabilistic modeling, we identify several classes of algorithmic strategies that can supply starting points to analysts, and thus help to discover interesting multi-relational patterns from datasets. To actually mine for interesting multi-relational patterns, we represent the multi-relational patterns as dense and well-connected chains of biclusters over multiple relations, and model the discrete data by the maximum entropy principle, such that in a statistically well-founded way we can gauge the surprisingness of a discovered bicluster chain with respect to what we already know. We design an algorithm for approximating the most informative multi-relational patterns, and provide strategies to incrementally organize discovered patterns into the background model. We illustrate how our method is adept at discovering the hidden plot in multiple synthetic and real-world intelligence analysis datasets. Our approach naturally generalizes traditional attribute-based maximum entropy models for single relations, and further supports iterative, human-in-the-loop, knowledge discovery. To build a generative model for multivariate categorical data, we apply the maximum entropy principle to propose a categorical maximum entropy model such that in a statistically well-founded way we can optimally use given prior information about the data, and are unbiased otherwise. Generally, inferring the maximum entropy model could be infeasible in practice. Here, we leverage the structure of the categorical data space to design an efficient model inference algorithm to estimate the categorical maximum entropy model, and we demonstrate how the proposed model is adept at estimating underlying data distributions. We evaluate this approach against both simulated data and US census datasets, and demonstrate its feasibility using an epidemic simulation application. Modeling data with multivariate count responses is a challenging problem due to the discrete nature of the responses. Existing methods for univariate count responses cannot be easily extended to the multivariate case since the dependency among multiple responses needs to be properly accounted for. To model multivariate data with multiple count responses, we propose a novel multivariate Poisson log-normal model (MVPLN). By simultaneously estimating the regression coefficients and inverse covariance matrix over the latent variables with an efficient Monte Carlo EM algorithm, the proposed model takes advantages of association among multiple count responses to improve the model prediction accuracy. Simulation studies and applications to real world data are conducted to systematically evaluate the performance of the proposed method in comparison with conventional methods. / Ph. D. / In this decade of big data, massive data of various types are generated every day from different research areas and industry sectors. Among all these types of data, text data, i.e. text documents, are important to many research and real world applications. One challenge faced when analyzing massive text data is which documents we should investigate first to initialize the analysis and how to identify stories and plots, if any, that hide inside the massive text documents. For example, in intelligence analysis, when analyzing intelligence documents, some common questions that analysts ask are ‘How is a suspect connected to the passenger manifest on this flight?’ and ‘How do distributed terrorist cells interface with each other?’. This is a crucial task so called storytelling. In the first half of this dissertation, we will study this problem and design mathematical models and computer algorithms to automatically identify useful information from text data to help analysts to discover hidden stories and plots from massive text documents. We also incorporate visual analytics techniques and design a visualization system to support human-in-the-loop exploratory data analysis so that analysts could interact with the algorithms and models iteratively to investigate given datasets. In the second half of this dissertation, we study two problems that arise from the domain of public health. When epidemic of certain disease happens, e.g. flu seasons, public health officials need to make certain policies in advance to prevent or alleviate the epidemic. A data-driven approach would be to make such public health policies using simulation results and predictions based on historical data. One problem usually faced in epidemic simulation is that researchers would like to run simulations with real-world data so that the simulation results can be close to real-world scenarios but at the same time protect the private information of individuals. To solve this problem, we design and implement a mathematical model that could generate realistic sythetic population using U.S. Census Survey to help conduct the epidemic simulation. Using flus as an example, we also propose a mathematical model to study associations between different types of flus with the information collected from social media, like Twitter. We believe that identifying such associations between different types of flus will help officials to make appropriate public health policies.

過去國內之抵押貸款提前清償與逾期還款之相關研究，在實證研究上最主要利用邏輯斯迴歸或是比例轉機模型( Proportional hazard model )分析影響一般住宅抵押貸款人提前清償與逾期還款之因素，並估計一般住宅抵押貸款人提前清償之機率。本文選擇採用研究抵押貸款時，國內未曾使用之卜瓦松迴歸( Poisson regression model )來估計比例轉機模型假設下影響提前清償與違約變數之參數，以研究影響抵押貸款借款人之提前償還與違約因素。 本研究結合比例轉機模型與卜瓦松迴歸模型，目的在結合兩模型之優點，在處理時間相依之共變數效率提高，並且在處理多重時間尺度的方程式較偏最大概似估計法直接，以得到較佳的研究成果。另外，過去國內提前清償與違約之文獻中並未加入利率走勢之變數，本研究加入再融資利率對31∼90天期商業本票利率之比率與再融資利率波動性兩變數，以考慮利率走勢對貸款者提前清償及違約行為之影響。 模型中的解釋變數包括地區、季節、抵押貸款年齡、貸款成數、貸款人年齡、性別、婚姻狀況、教育程度、職業、屋齡、房屋坪數、所得、貸款金額、月付額對薪資比、再融資利率/31∼90天期商業本票利率、再融資利率波動性等十六項。實證結果在提前清償部份，顯著正向之變數有貸款年齡、屋齡、房屋坪數、所得、月付額與薪資比，顯著負向之變數包括季節、再融資利率對31∼90天期商業本票利率之比率、貸款金額。在違約部份，顯著正向之變數包括貸款年齡、貸款成數、年齡、所得、月付額與薪資比、再融資利率對31∼90天期商業本票利率之比率；顯著負向之變數包括季節、教育程度及貸款金額。

Israël est l’un des pays développés les plus féconds dans le monde et maintient un taux de fécondité stable depuis 1995. Il a échappé à la chute spectaculaire de la fécondité qui a été observée dans la plupart des pays occidentaux. Le taux de fécondité était de 2,96 enfants par femme en 2009 (Statistical Abstract of Israel, 2010, tableau 3.14). Le maintien d’une si forte fécondité pourrait être dû à l’immigration et à la “guerre démographique” qui sévit entre les différentes communautés vivant dans le pays (Sardon, 2006). Toutefois, on observe une différence significative entre les niveaux de fécondité des juifs d’Israël et de Cisjordanie depuis plusieurs années. Les études qui portent sur la fécondité en Israël sont faites au niveau national, ce qui ne fournit aucune explication sur cette différence. Pour ces raisons, l’étude de la fécondité en Israël mérite une attention particulière. Ce projet vise à identifier les différents facteurs qui ont une incidence sur la fécondité des femmes juives vivant en Israël et en Cisjordanie. Il contribuera à une meilleure compréhension des comportements liés à la fécondité de la population juive de la Cisjordanie et peut fournir des indices sur les mécanismes complexes qui régissent les relations entre Juifs et Arabes dans les territoires occupés. Grâce aux données recueillies dans l’Enquête sociale générale de 2004 d’Israël,des analyses descriptives et explicatives ont été produites. Dans un premier temps, les facteurs qui ont un impact sur la fécondité dans chaque région ont été déterminés et par la suite, une analyse de l’importance de ces facteur sur la fécondité a été produite. Le nombre d’enfants nés de femmes âgées de 20 à 55 ans constitue la variable d’intérêt et les variables explicatives retenues sont les suivantes: religiosité, éducation, revenu familial mensuel, statut d’emploi, pays d’origine, âge et état matrimonial. Cette étude a montré que les femmes juives qui résident en Cisjordanie ont un nombre prévu d’enfants de 13% supérieur à celui des femmes juives qui résident en Israël lorsque l’on contrôle toutes les variables. Il est notamment montré que la religion joue un rôle important dans l’explication de la forte fécondité des femmes juives dans les deux régions, mais son impact est plus important en Israël. L’éducation joue également un rôle important dans la réduction du nombre prévu d’enfants, en particulier en Cisjordanie. Tous ces facteurs contribuent à expliquer les différents niveaux de fécondité dans les deux régions, mais l’étude montre que ces facteurs ne permettent pas une explication exhaustive de la forte fécondité en Israël et en Cisjordanie. D’autres forces qui ne sont pas mesurables doivent avoir une incidence sur la fécondité telles que le nationalisme ou la laïcisation, par exemple. / Israel is one of the most fertile developed countries in the world and has had a stable fertility rate since 1995. The country avoided the dramatic fall in fertility that has been observed in most Western countries. The fertility rate was of 2.96 children per woman in 2009 (Statistical Abstract of Israel, 2010, table 3.14). Maintaining such a high fertility level could be due to immigration and the “demographic war” between the different communities living in the country (Sardon, 2006). However, a significant difference between the levels of fertility of the jewish population of Israel and the West Bank has been observed for several years. In the literature, studies of fertility in Israel are conducted at a national level, which neither reveals nor explains the difference. Accordingly, Israel’s high fertility deserves a particular attention. This project aims to identify the different factors that affect the fertility of Jewish women living in Israel and in the West Bank. It will contribute to a better understanding of the fertility behavior of the Jewish population of the West Bank and may shed light on the complex mechanisms that govern the relations between Jews and Arabs in the Occupied Territories. With data collected in the General Social Survey of Israel of 2004, descriptive and explanatory analyses were produced. In the first part, factors influencing fertility in each region have been determined and an analysis of the importance of each factor on fertility was conducted in the second part. The outcome of interest is the number of children ever born to women aged 20 to 55 and the independent variables are: religiosity, education, monthly family income, employment status, country of origin, age and marital status. This study showed that Jewish women residing in the West Bank have an expected number of children 13% higher than their counterparts residing in Israel. It is also shown that the intensity of religious interest plays an important role in explaining the high fertility of Jewish women in both regions but its impact is more important in Israel. Education also plays an important role in reducing the expected number of children, especially in the West Bank. All of these factors contribute to explaining the different fertility levels in the two regions but the study shows that these factors do not provide an exhaustive explanation of higher fertility in the West Bank. There must be other forces that have an impact on fertility but which are not measurable such as nationalism or secularization, for example.

La méthode que nous présentons pour modéliser des données dites de "comptage" ou données de Poisson est basée sur la procédure nommée Modélisation multi-niveau et interactive de la régression de Poisson (PRIMM) développée par Christiansen et Morris (1997). Dans la méthode PRIMM, la régression de Poisson ne comprend que des effets fixes tandis que notre modèle intègre en plus des effets aléatoires. De même que Christiansen et Morris (1997), le modèle étudié consiste à faire de l'inférence basée sur des approximations analytiques des distributions a posteriori des paramètres, évitant ainsi d'utiliser des méthodes computationnelles comme les méthodes de Monte Carlo par chaînes de Markov (MCMC). Les approximations sont basées sur la méthode de Laplace et la théorie asymptotique liée à l'approximation normale pour les lois a posteriori. L'estimation des paramètres de la régression de Poisson est faite par la maximisation de leur densité a posteriori via l'algorithme de Newton-Raphson. Cette étude détermine également les deux premiers moments a posteriori des paramètres de la loi de Poisson dont la distribution a posteriori de chacun d'eux est approximativement une loi gamma. Des applications sur deux exemples de données ont permis de vérifier que ce modèle peut être considéré dans une certaine mesure comme une généralisation de la méthode PRIMM. En effet, le modèle s'applique aussi bien aux données de Poisson non stratifiées qu'aux données stratifiées; et dans ce dernier cas, il comporte non seulement des effets fixes mais aussi des effets aléatoires liés aux strates. Enfin, le modèle est appliqué aux données relatives à plusieurs types d'effets indésirables observés chez les participants d'un essai clinique impliquant un vaccin quadrivalent contre la rougeole, les oreillons, la rub\'eole et la varicelle. La régression de Poisson comprend l'effet fixe correspondant à la variable traitement/contrôle, ainsi que des effets aléatoires liés aux systèmes biologiques du corps humain auxquels sont attribués les effets indésirables considérés. / We propose a method for analysing count or Poisson data based on the procedure called Poisson Regression Interactive Multilevel Modeling (PRIMM) introduced by Christiansen and Morris (1997). The Poisson regression in the PRIMM method has fixed effects only, whereas our model incorporates random effects. As well as Christiansen and Morris (1997), the model studied aims at doing inference based on adequate analytical approximations of posterior distributions of the parameters. This avoids the use of computationally expensive methods such as Markov chain Monte Carlo (MCMC) methods. The approximations are based on the Laplace's method and asymptotic theory. Estimates of Poisson mixed effects regression parameters are obtained through the maximization of their joint posterior density via the Newton-Raphson algorithm. This study also provides the first two posterior moments of the Poisson parameters involved. The posterior distributon of these parameters is approximated by a gamma distribution. Applications to two datasets show that our model can be somehow considered as a generalization of the PRIMM method since it also allows clustered count data. Finally, the model is applied to data involving many types of adverse events recorded by the participants of a drug clinical trial which involved a quadrivalent vaccine containing measles, mumps, rubella and varicella. The Poisson regression incorporates the fixed effect corresponding to the covariate treatment/control as well as a random effect associated with the biological system of the body affected by the adverse events.

This study proposes a social and emotional disclosure model for understanding the mechanism that explains sharing intimate information on social media (Twitter). Previous research has indicated that some aspects of social, emotional, and attitudinal information processing are involved in disclosure of intimate information. However, these factors have been considered in isolation. This study proposes and tests a theoretically grounded model that brings all of these factors together by combining individual and group social media behaviors and online information processing in the realm of online social movements. The core explanatory model considers the impact of peer response, emotional evaluation, personal relevance, issue orientation, and motivation to post online on intimate information disclosure online. A path analysis building on four Poisson multiple regressions conducted on 28,629 #MeToo tweets evaluates the relationships proposed in the explanatory model. Results indicate that emotional evaluation and motivation to post online have direct, positive impacts on online disclosure. Other factors such as peer response, issue orientation, and personal relevance have negative direct relationships with online disclosure. Motivation to post online mediates the effects of emotional evaluation, issue orientation, and personal relevance on online disclosure while issue orientation mediates the effect of personal relevance on motivation to post online. This study offers findings that have use for practitioners interested in hashtag virality and to social media users interested in social influence and online information sharing.

Mapping of mortality rates has been a valuable public health tool. We describe novel Bayesian methods for constructing maps which do not depend on a post stratification of the estimated rates. We also construct posterior modal maps rather than posterior mean maps. Our methods are illustrated using mortality data from chronic obstructive pulmonary diseases (COPD) in the continental United States. Poisson regression models have attracted much attention in the scientific community for their superiority in modeling rare events (including mortality counts from COPD). Christiansen and Morris (JASA 1997) described a hierarchical Bayesian model for heterogeneous Poisson counts under the exchangeability assumption. We extend this model to include latent classes (groups of similar Poisson rates unknown to an investigator). Also, it is standard practice to construct maps using quantiles (e.g., quintiles) of the estimated mortality rates. For example, based on quintiles, the mortality rates are cut into 5 equal size groups, each containing $20\%$ of the data, and a different color is applied to each of them on the map. A potential problem is that, this method assumes an equal number of data in each group, but this is often not the case. The latent class model produces a method to construct maps without using quantiles, providing a more natural representation of the colors. Typically, for rare events, the posterior densities of the rates are skewed, making the posterior mean map inappropriate and inaccurate. Thus, although it is standard practice to present the posterior mean maps, we also develop a method to provide the joint posterior modal map (i.e., the map with the highest posterior probability over the ensemble). For the COPD data, collected 1988-1992 over 798 health service areas, we use Markov chain Monte Carlo methods to fit the model, and an output analysis is used to construct the new maps.

Estudos envolvendo o meio ambiente estão sendo cada vez mais desenvolvidos devido ao fato dos níveis de poluição e das mudanças climáticas estarem causando a degradação da qualidade do ar e dos reservatórios de água de maneira alarmante nos últimos anos, comprometendo sobretudo, a qualidade de vida do ser humano. Dado que estes fatores são preponderantes nos agravos e complicações respiratórias dos indivíduos, buscou-se compreender com este estudo a relação entre as condições atmosféricas e os problemas respiratórios nos residentes do município de Ribeirão Preto, interior de São Paulo, onde há um elevado número de focos de queimadas nos períodos de estiagem e, consequentemente, altas concentrações de poluentes, como o material particulado. Considerando os dados mensais de contagem de inalações/nebulizações, foram assumidos diferentes modelos de regressão de Poisson na presença de um fator aleatório que captura a variabilidade extra-Poisson entre as contagens. A análise dos dados foi feita sob enfoque Bayesiano, utilizando métodos de simulação MCMC (Monte Carlo em Cadeias de Markov) para obter os sumários a posteriori de interesse. / Many studies involving the environment are being developed in the last years due to the fact that the levels of pollution and climate changes are causing the degradation of air quality and water reservoirs at an alarming rate in recent years, with great consequences for the quality of life of the population. Since these factors are prevalent in respiratory disorders and complications of the health for the individuals, we intended to understand from this study the relationship between weather conditions and respiratory problems for the residents of the municipality of Ribeirão Preto, São Paulo, which has a high number of outbreaks of fires in drought periods and, consequently, high concentrations of pollutants such as particulate matter. Considering the monthly count of inhalations / nebulizations, we assumed different Poisson regression models in the presence of a random factor that captures the extra-Poisson variability between the counts. The data analysis was performed under a Bayesian approach using MCMC simulation methods (Markov Chain Monte Carlo) to get the posterior summaries of interest.

Estudos envolvendo o meio ambiente estão sendo cada vez mais desenvolvidos devido ao fato dos níveis de poluição e das mudanças climáticas estarem causando a degradação da qualidade do ar e dos reservatórios de água de maneira alarmante nos últimos anos, comprometendo sobretudo, a qualidade de vida do ser humano. Dado que estes fatores são preponderantes nos agravos e complicações respiratórias dos indivíduos, buscou-se compreender com este estudo a relação entre as condições atmosféricas e os problemas respiratórios nos residentes do município de Ribeirão Preto, interior de São Paulo, onde há um elevado número de focos de queimadas nos períodos de estiagem e, consequentemente, altas concentrações de poluentes, como o material particulado. Considerando os dados mensais de contagem de inalações/nebulizações, foram assumidos diferentes modelos de regressão de Poisson na presença de um fator aleatório que captura a variabilidade extra-Poisson entre as contagens. A análise dos dados foi feita sob enfoque Bayesiano, utilizando métodos de simulação MCMC (Monte Carlo em Cadeias de Markov) para obter os sumários a posteriori de interesse. / Many studies involving the environment are being developed in the last years due to the fact that the levels of pollution and climate changes are causing the degradation of air quality and water reservoirs at an alarming rate in recent years, with great consequences for the quality of life of the population. Since these factors are prevalent in respiratory disorders and complications of the health for the individuals, we intended to understand from this study the relationship between weather conditions and respiratory problems for the residents of the municipality of Ribeirão Preto, São Paulo, which has a high number of outbreaks of fires in drought periods and, consequently, high concentrations of pollutants such as particulate matter. Considering the monthly count of inhalations / nebulizations, we assumed different Poisson regression models in the presence of a random factor that captures the extra-Poisson variability between the counts. The data analysis was performed under a Bayesian approach using MCMC simulation methods (Markov Chain Monte Carlo) to get the posterior summaries of interest.