nÃo hà / Nosso principal objetivo neste estudo à investigar o papel da heterogeneidade na atualizaÃÃo, depois de um choque de informaÃÃo, do risco subjetivo sobre vitimizaÃÃo de homicÃdio. Nesse sentido, os dados utilizados neste trabalho tambÃm atestam a superestimaÃÃo do crime encontrada na literatura. A novidade à que os entrevistados receberam um choque de informaÃÃo que consiste na taxa oficial de homicÃdios, mas a grande maioria deles mantÃm a mesma percepÃÃo inicial. Ao propor um modelo de Update Bayesiano permitindo que nenhuma atualizaÃÃo fosse realizada, dois modelos foram desenvolvidos: um Tobit modificado e um modelo Hurdle de dois nÃveis. Assim como em estudos anteriores, nossos resultados mostraram que poderÃamos prosseguir com uma abordagem de Update Bayesiano. Ainda, quanto mais altas as respostas iniciais eram definidas, mais propensos os indivÃduos estavam em proceder uma mudanÃa de percepÃÃo. AlÃm disso, fundamentalmente, pudemos racionalizar a decisÃo de nÃo revisar as respostas seguindo um argumento de qualidade/credibilidade da informaÃÃo percebida. Descobrimos que os participantes mais velhos e as mulheres sÃo mais relutantes nÃo apenas em alterar as respostas iniciais, mas tambÃm na escolha do nÃvel da nova resposta, em caso de mudanÃa. Outra conclusÃo feita foi que o nÃvel educacional dos entrevistados era insignificante em nosso exercÃcio. De fato, o nÃvel educacional do entrevistador teve um papel fundamental em ambas decisÃes de mudanÃa e magnitude de revisÃo. Finalmente, nossos resultados tambÃm levantaram fortes evidÃncias sobre aspectos de homofilia. A ocorrÃncia de uma correspondÃncia em gÃnero entre entrevistadores e entrevistados teve o maior impacto sobre a decisÃo de mudar e na magnitude da atualizaÃÃo neste estudo. / Our main purpose in this study is to investigate the role of heterogeneity into the update of subjective homicide victimization risk after an informational shock. In this sense, the data used here also attests the crime overestimation found in the literature. The novelty is that our respondents faced an informational shock consisting in the official homicide rate, but the vast majority of them keeps the same initial perception. In proposing a Bayesian Update model allowing that no update takes place, two models were developed: a modified Tobit and a two-tiered Hurdle model. In accordance with previous papers, our results showed that we could proceed with a Bayesian Update approach. Also, the higher initial responses are set, more likely individuals are in proceeding a change in perceptions. Furthermore, fundamentally, we could rationalize a non-updating decision following a perceived informational quality/credibility argument. We found that older participants and females are more reluctant not only to change initial responses, but also to choose the level of the new response, in case of an update. In addition, respondentsâ level of education was insignificant in our exercise. In fact, interviewersâ level of education had a key role in both the changing and updating magnitude decisions. Finally, our results also raised strong evidence on homophily aspects. The occurance of a matching in gender between interviewers and interviewees had a major impact on the decision to change and in the magnitude of the update in this study.

The American horseshoe crab (HSC), Limulus polyphemus, is one of four species of horseshoe crabs found throughout the world, and the only one found in North America. It is an economically and ecologically important species throughout its native range from Maine to the Yucatan Peninsula. Harvested for fertilizer and livestock feed in the 19th century, the species is now harvested as bait for whelk and eel fisheries, and for their blood by the biomedical industry. The Atlantic States Marine Fisheries Commission (ASMFC) started to formally manage HSC in 1998 with its Interstate Fisheries Management Plan (IFMP). Unique emphasis and harvest limits have been placed on the Delaware Bay stock, as it is commercially exploited and a critical food source for the threatened red knot, Calidris canutus rufa. Previously, estimates of relative and total abundance of HSC in the Delaware Bay area were based on a design-based approach using a stratified random sampling design. In Chapter 1 of this work, I developed hurdle models for each of the six HSC demographic groups to standardize catch-per-unit-effort (CPUE) and estimate relative abundance using a model-based approach. It was determined that while the two approaches resulted in mostly convergent estimates of relative abundance, external factors such as month, time-of-day, and average depth have major effects on the observed CPUE of all demographic groups. Chapter 2 involved the development of hurdle models for the three species of bycatch frequently caught in our trawls, channeled whelk (Busysotypus canaliculatus), knobbed whelk (Busycon carica), and summer flounder (Paralichthys dentatus). It was found that channeled whelk relative abundance has been at a historical low since 2016, while summer flounder has been at a consistent high. Recent estimates of knobbed whelk relative abundance have been less variable than previously seen, with estimates since 2016 being similar to those seen before 2012. These results provide the first estimates for whelk population trends in the mid-Atlantic region and add to the growing knowledge of summer flounder relative abundance in the area. In Chapter 3, I applied the hurdle models developed in Chapter 1 to estimate the total abundance of HSC in the Delaware Bay area. For this work, I developed two spatio-temporal variograms to estimate bottom temperature and bottom salinity at unmeasured cells per month in the time series. The results showed that night estimates of total abundance were consistently higher than daytime estimates, and estimates from September or November resulted in the highest estimated catch for all demographic groups. The results suggest that when comparing September model-based estimates at night to those of the design-based approach, nearly a third of all previous design-based estimates significantly underestimated the total abundance of HSC in the Delaware Bay area. This result suggests that the ASMFC can recommend increased harvest limits for mature individuals if that action aligns with the goals of their adaptive resource management (ARM) framework. / Master of Science / The American horseshoe crab (HSC), Limulus polyphemus, is one of four species of horseshoe crabs found throughout the world, and the only one found in North America. It is an economically and ecologically important species throughout its native range from Maine to the Yucatan Peninsula. Harvested by the millions in the 19th century, the species is still harvested as bait for whelk and eel fisheries, and for their blood by the biomedical industry, on the order of hundreds of thousands. Formal management of HSC by the Atlantic States Marine Fisheries Commission (ASMFC) began in 1998, and allowable catch and landings have decreased since 1999. A strong focus has been placed on the Delaware Bay population, as it is the center of HSC abundance and provides a critical food source for the threatened red knot, Calidris canutus rufa, while also being commercially exploited. To effectively manage the species, it is important to accurately estimate relative and total abundance so that proper harvest limits can be set. Previously, estimates of relative and total abundance of HSC in the Delaware Bay area were based on a design-based approach using a stratified random sampling design. In Chapter 1 of this work, I developed hurdle models (a generalized linear model that models the probability of observations and the observed positive counts using two separate regression models that are then combined) for each of the six HSC demographic groups to remove the effect of external factors (year, latitude, longitude, depth [inshore/offshore], topography, average trawl depth, time-of-day, month, bottom temperature, bottom salinity, and distance from shore) on our observed catch-per-unit-effort (CPUE) and estimate relative abundance using a model-based approach. It was determined that while the two approaches resulted in mostly convergent estimates of relative abundance, factors like month, time-of-day, and average depth had major effects on the observed CPUE of all demographic groups. Chapter 2 involved developing similar hurdle models for three species of bycatch frequently caught in our trawls, i.e., channeled whelk (Busysotypus canaliculatus), knobbed whelk (Busycon carica), and summer flounder (Paralichthys dentatus). It was found that channeled whelk relative abundance has been at a historical low since 2016, while summer flounder has been at a consistent high. Recent estimates of knobbed whelk relative abundance have been less variable than previously seen, with estimates since 2016 being similar to those seen before 2012. These results provide the first estimates for whelk population trends in the mid-Atlantic region and add to the growing knowledge of summer flounder relative abundance in the area. In Chapter 3, I applied the hurdle models developed in Chapter 1 to estimate the total abundance of HSC in the Delaware Bay area. To do this, the bottom temperature and salinity had to be estimated for each geographic cell. This was accomplished by developing two spatio-temporal variograms which allowed me to estimate either variable at an unmeasured point and time based on its spatial and temporal distance from a measured value in a process known as spatio-temporal kriging. The results showed that night estimates of total abundance were consistently higher than daytime estimates and that estimates from September or November resulted in the highest estimated total abundance for all demographic groups. The results suggest that when using September model-based estimates at night to compare against the design-based approach, nearly a third of all previous design-based estimates significantly underestimated the total abundance of HSC in the Delaware Bay area. This outcome could justify ASMFC increasing recommended harvest limits for mature individuals if that action aligns with the goals of their adaptive resource management (ARM) framework.

Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro / Apesar dos progressos recentes do setor de saneamento básico no Brasil, muitas dificuldades ainda precisam ser superadas. A desigualdade no acesso a esses serviços bem como a fragmentação das políticas públicas e a ausência de marco regulatório podem ser citados como os principais desafios no setor. Este estudo tem por objetivo analisar os determinantes da cobertura de saneamento básico no Brasil, com base na Pesquisa Nacional de Saneamento Básico de 1989 e 2000 e nos Censos demográficos de 1991 e 2000 elaborados pelo Instituto Brasileiro de Geografia e Estatística (IBGE). A fim de estimar tanto os fatores que determinam a existência de esgotamento sanitário nos domicílios de um município (modelo de variável dependente limitada) quanto os fatores que indicam a proporção de domicílios cobertos por esgotamento sanitário (modelo Double-Hurdle), foram utilizadas variáveis de demanda, variáveis de oferta, variáveis institucionais e variáveis de controle na estimação de um modelo. Como resultado, observou-se, por exemplo, que a variável renda per capita municipal não é estatisticamente significativa no primeiro modelo mas o é no segundo, o que parece indicar, por um lado, que a existência de bolsões de pobreza é mais importante do que o nível médio de renda do município no caso da determinação da existência de esgotamento sanitário em um município mas, por outro lado, que a renda per capital é o principal determinante na expansão da rede de esgoto. Por fim, o estudo provou que avanços podem ser feitos no que tange os condicionantes políticos que parecem afetar de forma desproporcional e desigual os gastos dos municípios com esgotamento sanitários. / Despite some progress authorities have been promoting in the past few years, many difficulties remain in the Brazilian basic sanitation sector. The inequality in access to those public services, the non-coordinated public policies and the absence of an efficient regulatory framework are the main challenges to the sector. This study analyses the factors behind the basic sanitation coverage in Brazil, based on the Pesquisa Nacional de Saneamento Básico 1989/2000 (1989 and 2000 National Survey of Basic Sanitation) and the Censo Demográfico 1991/2000 (1991 and 2000 Demographic Census), both published by the Instituto Brasileiro de Geografia e Estatística (IBGE). Two models were built in a way to estimate the factors that determine the installation of sanitary sewer services in Brazilian houses and the factors that explain the ratio of houses covered by sanitary sewer services in Brazilian cities - the first, a limited dependent variable model, and the second a Double-Hurdle model, both using demand, supply, institutional and control variables. An important result, for example, is that the variable income per capita is not statistically significant for the first model but it is for the second, which seems to indicate, on one hand, that the fact that exists many bolsões de pobreza (communities of concentrated poverty) are more important than the city basic level of income in the case of determining the installation of sanitary sewer services in Brazilian houses but, on the other hand, that income per capita is the main factor in the sewer services expansion. At last, progress can be made by studding more carefully the political effects that seems to affect disproportionally and unequally cities expenses with sewer services.

Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro / Apesar dos progressos recentes do setor de saneamento básico no Brasil, muitas dificuldades ainda precisam ser superadas. A desigualdade no acesso a esses serviços bem como a fragmentação das políticas públicas e a ausência de marco regulatório podem ser citados como os principais desafios no setor. Este estudo tem por objetivo analisar os determinantes da cobertura de saneamento básico no Brasil, com base na Pesquisa Nacional de Saneamento Básico de 1989 e 2000 e nos Censos demográficos de 1991 e 2000 elaborados pelo Instituto Brasileiro de Geografia e Estatística (IBGE). A fim de estimar tanto os fatores que determinam a existência de esgotamento sanitário nos domicílios de um município (modelo de variável dependente limitada) quanto os fatores que indicam a proporção de domicílios cobertos por esgotamento sanitário (modelo Double-Hurdle), foram utilizadas variáveis de demanda, variáveis de oferta, variáveis institucionais e variáveis de controle na estimação de um modelo. Como resultado, observou-se, por exemplo, que a variável renda per capita municipal não é estatisticamente significativa no primeiro modelo mas o é no segundo, o que parece indicar, por um lado, que a existência de bolsões de pobreza é mais importante do que o nível médio de renda do município no caso da determinação da existência de esgotamento sanitário em um município mas, por outro lado, que a renda per capital é o principal determinante na expansão da rede de esgoto. Por fim, o estudo provou que avanços podem ser feitos no que tange os condicionantes políticos que parecem afetar de forma desproporcional e desigual os gastos dos municípios com esgotamento sanitários. / Despite some progress authorities have been promoting in the past few years, many difficulties remain in the Brazilian basic sanitation sector. The inequality in access to those public services, the non-coordinated public policies and the absence of an efficient regulatory framework are the main challenges to the sector. This study analyses the factors behind the basic sanitation coverage in Brazil, based on the Pesquisa Nacional de Saneamento Básico 1989/2000 (1989 and 2000 National Survey of Basic Sanitation) and the Censo Demográfico 1991/2000 (1991 and 2000 Demographic Census), both published by the Instituto Brasileiro de Geografia e Estatística (IBGE). Two models were built in a way to estimate the factors that determine the installation of sanitary sewer services in Brazilian houses and the factors that explain the ratio of houses covered by sanitary sewer services in Brazilian cities - the first, a limited dependent variable model, and the second a Double-Hurdle model, both using demand, supply, institutional and control variables. An important result, for example, is that the variable income per capita is not statistically significant for the first model but it is for the second, which seems to indicate, on one hand, that the fact that exists many bolsões de pobreza (communities of concentrated poverty) are more important than the city basic level of income in the case of determining the installation of sanitary sewer services in Brazilian houses but, on the other hand, that income per capita is the main factor in the sewer services expansion. At last, progress can be made by studding more carefully the political effects that seems to affect disproportionally and unequally cities expenses with sewer services.

Life insurance policies are not equally profitable is sense of expected value. In practice, profitability is an output of complex cash flow models, which need utilizing special systems and the run time of such calculation can be significant if number of policies is high. Therefore we consider variables, which change most frequently, stimulate the profitability model with several values of these variables and then we search for a regression model to explain the changes. We apply Gamma regression on the data. But what if there exist some policies which are negative? Then we determine these policies with logistic regression applied on data censored to the binary form. Loss of these policies is modelled using symmetrical Gamma model. These three models, when considered together, can be viewed as a single model, which is a generalization of the well known zero inflated count model. The most interesting part of inference in such model is diagnostics. We show that the basic types of residuals - Pearson, deviance and quantile - can be defined. We also build an ordinary linear model and we compare utility of these two approaches. While building models, we meet various statistical issues like dimension reduction of yield curve or dispersion proportional to sum insured. 1

Cette thèse de doctorat examine les principales défaillances du marché du travail qui entraînent que les travailleurs n’arrivent pas à obtenir la pleine rémunération potentielle qui corresponde à leur capital humain. Il y a « inefficacité salariale » lorsque le salaire obtenu est inférieur au maximum atteignable. Dans un tel cas, les salariés reçoivent un salaire injuste par rapport au capital humain disposé. Cela décourage à investir dans son capital humain ce qui aura tendance à réduire la productivité totale, à affaiblir la compétitivité et à nuire à la croissance économique du pays. La contribution que nous apportons à travers ce travail est de trois ordres. Dans un premier temps, nous proposons un nouveau regard par rapport à l’intégration des jeunes sur le marché du travail. Plutôt que d’examiner si les individus obtiennent un contrat de travail stable ou non, nous adoptons une approche qui s’intéresse à la qualité de l’appariement « emploi – compétences » de jeunes entrant fraîchement dans la vie active. Nos travaux fournissent des résultats empiriques qui mettent en évidence les différentes théories de recherche d’emploi. Dans un second temps, en raison du considérable défi que représente l'identification et l'évaluation des pratiques discriminatoires sur le marché du travail, nous proposons une approche innovatrice et efficace pour examiner le phénomène du plafond de verre (barrière invisible à l’accès des postes de décision mieux rémunérés). Dans la dernière partie de cette thèse, nous proposons un modèle économétrique théorique qui améliore la correction du problème de biais de sélection pour les modèles de frontière stochastique. / This doctoral thesis addresses issues related to employees’ imperfect information on the labour market and discrimination, generally all direct consequences of labour underpayment or “earnings inefficiency”. Workers are in a situation of earnings inefficiency when they do not receive the full potential remuneration corresponding to their human capital endowment: unfair pay for greater stock of human capital. This situation is problematic from a policy-makers point of view as it could weaken work incentives, discourage investments in human capital, and harm economic growth and competitiveness. It could also widen inequality within the society and contribute to the increase in relative poverty.The contribution we make through this work is threefold. First, we examine the integration of young people into the labour market from a new angle. Instead of examining whether individuals obtain stable employment or not, we use an approach that focuses on the quality of the job matching for young people entering the workforce and lacking labour market information. This first chapter provides empirical evidence on job search theories. In a second chapter, because of the considerable challenge of identifying and assessing discriminatory practices in the labour market, we propose an innovative and effective approach to examine the phenomenon of the glass ceiling (an invisible barrier to management positions associated with higher earnings). Finally, the last part of this doctoral thesis is devoted to improving the econometric approach we use. We propose a theoretical econometric model that improves correction for sample selection bias with stochastic frontier models.