The last decades have seen a rapid increase of sustainable investing, also known as ESG (Environmental, Social and Governance) investing. There has also been an increasing body of academic literature devoted to whether investors can gain any financial benefits from taking ESG under consideration. Previous literature of portfolio performance in terms of risk-adjusted returns has given much of its attention to best-in-class approaches, which is a strategy that selects top performers in ESG within a sector or industry. The purpose of this study is foremost to investigate a best-effort approach to ESG investing, which is a strategy that focuses on the top improvers in ESG. The purpose is further to compare this with a best-in-class approach, since the findings from earlier studies of this strategy still are inconsistent. The region chosen to perform this study in is the Eurozone. Several theories that have implications for portfolio studies and abnormal returns are taken under consideration in relation to the study and its findings. This includes the efficient market hypothesis, the adaptive market hypothesis and modern portfolio theory. The theoretical framework also cover asset-pricing models and the notions of risk-adjusted returns. A quantitative study with a deductive approach are used to form portfolios, with a Eurozone index as the investable universe. Best-effort and best-in-class portfolios as well as difference portfolios of the two approaches are created, based on ESG data and different cut-off rates for portfolio inclusion. As for risk-adjusted performance measure, the Carhart four-factor model are used. The overall results are mostly insignificant findings in terms of abnormal returns. However, three best-effort portfolios based on the top ESG improvers show significant positive abnormal returns. These findings are strongest for the environmental and social factor. As for the best-in-class approach, only the governance portfolios provided weakly significant results in terms of abnormal returns. Further, the study is not able to significantly distinguish between a best-effort and a best-in-class approach when it comes to risk-adjusted performance. The exception is the environmental factor based on the top performers in each approach, where the best-effort portfolio outperforms the best-in-class portfolio. Finally, none of the portfolios provided significant negative risk-adjusted returns. This can at least be considered as good news for ESG investing, since it indicates that investors do not have to sacrifice risk-adjusted returns in order to invest in a more sustainable way.

This paper aims to examine if firms listed on Nasdaq Stockholm with dividend exposure yield higher risk-adjusted returns than firms without dividend exposure. Using a data set consisting of observations between 2000-2017 we test the difference in mean risk-adjusted return, measured by the Sharpe ratio, between securities with different levels of dividend exposure. We divide our sample into portfolios, categorized in the first stage independently of investment style, size and book-to-market ratio, and in the second stage on dividend exposure, that are regrouped annually. We measure the performance in terms of the geometric mean monthly returns, the risk as standard deviation of returns and the risk-adjusted performance measured with the Sharpe ratio. Following our empirical study, we find indications of a value effect in the Swedish capital market and draw upon three main conclusions. First, for all but one portfolio, the risk decreases with an increased degree of dividend exposure. Second, securities with high-dividend exposure tend to yield higher risk-adjusted returns relative to securities with no-dividend exposure. Third, the effect of dividend exposure on risk-adjusted performance appears to be most significant on mid firms and growth firms

A principal motivação desse estudo é a análise da quantidade de sangue recebido em transfusão (padronizada pelo peso) por crianças com problemas hepáticos. Essa quantidade apresenta distribuição assimétrica, além de valores iguais a zero para as crianças que não receberam transfusão. Os modelos lineares generalizados, usuais para variáveis positivas, não permitem a inclusão de zeros. Para os dados positivos, foram ajustados tais modelos com distribuição gama e normal inversa. Também foi considerado o modelo log-normal. A análise de resíduos padronizados indicou heterocedasticidade, e portanto a variabilidade extra foi modelada utilizando a classe de modelos GAMLSS. A terceira abordagem consiste em modelos baseados na mistura de zeros e distribuições para valores positivos, incluídos recentemente na família dos modelos GAMLSS. Estes aliam a distribuição assimétrica para os dados positivos e a probabilidade da ocorrência de zeros. Na análise dos dados de transfusão, observa-se que a distribuição normal inversa apresentou melhor ajuste por acomodar dados com forte assimetria em relação às demais distribuições consideradas. Foram significativos os efeitos das variáveis explicativas Kasai (ocorrência de operação prévia) e PELD (nível de uma medida da gravidade do paciente com 4 níveis) assim como os efeitos de interação sobre a média e variabilidade da quantidade de sangue recebida. A possibilidade de acrescentar efeitos de variáveis explicativas para modelar o parâmetro de dispersão, permite que a variabilidade extra, além de sua dependência da média, seja melhor explicada e melhore o ajuste do modelo. A probabilidade de não receber transfusão depende de modo significativo somente de PELD. A proposta de um só modelo que alia a presença de zeros e diversas distribuições assimétricas facilita o ajuste dos dados e a análise de resíduos. Seus resultados são equivalentes à abordagem em que a ocorrência ou não de transfusão é analisada por meio de modelo logístico independente da modelagem dos dados positivos com distribuições assimétricas. / The main motivation of this study is to analyze the amount of blood transfusions received (by weight) of children with liver problems. This amount shows asymmetric distribution as well as present zero values for children who did not receive transfusions. The usual generalized linear models for positive variables do not allow the inclusion of zeros. For positive data, such models with gamma and inverse normal distributions were fitted in this study. Log-normal distribution was also considered. Analysis of the standardized residuals indicated heterocedasticity and therefore the extra variability was modelled using GAMLSS. The third approach consists of models based on a mixture of zeros and distributions for positive values, also recently included in the family of GAMLSS models. These models combine the asymmetric distribution of positive data and the probability of occurrence of zeros. In the data analysis of transfusion, the inverse normal distribution showed better goodness of fit to allow adjustment of data with greater asymmetry than the other distributions considered. The effects of the explanatory variables Kasai (occurrence of previous operation) and PELD level (a measure of the severity of the patient with 4 levels) and interaction effects on the mean and variability of the amount of blood received were signicant. The inclusion of explanatory variables to model the dispersion parameter, allows to model the extra variability, beyond its dependence on the average, and improves fitness of the model. The probability of not receiving transfusion depends signicantly only PELD. The proposal of a unified model that combines the presence of zeros and several asymmetric distributions greatly facilitates the fitness of the model and the evaluation of fitness. An advantage is the equivalence between this model and a separate logistic model to for the probability of the occurrence of transfusion and a model for the positive skewed data.

O presente estudo tem por objetivo investigar as características e implicações da divulgação voluntária das medidas de desempenho não-GAAP \"EBITDA\" e \"EBITDA Ajustado\" em relatórios financeiros preparados conforme IFRS no cenário corporativo brasileiro. A principal preocupação relacionada a divulgações voluntárias é se de fato estas informações garantem a qualidade do processo decisório dos usuários. A natureza do estudo é empírico-teórica com abordagem qualitativa e quantitativa. Os dados documentais foram extraídos de três fontes: Thomson Reuters; relatórios anuais e press releases disponibilizados no site de cada companhia; e formulários de referência disponibilizados no site da CVM. A amostra selecionada é a do índice IBrX 100 e os dados analisados são dos períodos trimestrais e anuais de 2014 e 2015. Para coleta das percepções sobre o uso e divulgação das métricas estudadas, aplica-se o questionário semiestruturado com sócios de firmas BigFour. Os principais resultados qualitativos sugerem, em linhas gerais: que o uso e divulgação das medidas estudadas tem ocorrido de forma ampla, consistente e regular; que 79% dos ajustes adicionais efetuados pelas companhias por meio do EBITDA Ajustado são consequência de princípios e regras contábeis vigentes conforme IFRS; que os ajustes adicionais mais comuns efetuados pelas companhias são Impairment, Provisões, Correção de Erros e Equivalência Patrimonial; e que o uso e divulgação das medidas investigadas se faz necessário uma vez que a contabilidade não é suficiente para prover aos seus usuários uma medida que forneça o desempenho isolado da atividade operacional de uma companhia. Pode-se concluir com base nos resultados quantitativos deste estudo que empresas de maior porte, que aderem a níveis de governança corporativa e que possuem receitas líquidas menores, estão mais propensas a divulgar as medidas \"EBITDA\" e \"EBITDA Ajustado\". As evidências advindas deste estudo podem ser úteis para colaborar com a discussão atual de órgãos reguladores e normatizadores ao apontar o papel informativo de medidas alternativas de mensuração, mas não deixando de alertar que estes números necessitam ser acompanhados e fiscalizados pelos entes e instituições cabíveis. / The present study aims to investigate the characteristics and implications of the voluntary disclosure of non-GAAP performance measures \"EBITDA\" and \"Adjusted EBITDA\" in financial reports prepared under IFRS in the Brazilian corporate scenario. The main concern related to voluntary disclosures is whether this information actually guarantees the quality of the user\'s decision-making process. The nature of the study is empirical-theoretical with a qualitative and quantitative approach. Documentary data were extracted from three sources: Thomson Reuters; annual reports and press releases made available on each company\'s website; and reference forms available on the CVM website. The selected sample is the IBrX 100 index and the data analyzed are from the quarterly and annual periods of 2014 and 2015. To collect the perceptions about the use and disclosure of the metrics studied, the semi-structured questionnaire is applied with partners of BigFour firms. The main qualitative results suggest, in general lines: that the use and disclosure of the measures studied has occurred in a broad, consistent and regular way; that 79% of the additional adjustments made by the companies through Adjusted EBITDA are a consequence of accounting principles and rules according to IFRS; that the most common additional adjustments made by the companies are Impairment, Provisions, Errors and Equity Method; and that the use and disclosure of the measures investigated becomes necessary since accounting is not sufficient to provide its users with a measure that provides the isolated performance of a company\'s operating activity. It can be concluded from the quantitative results of this study that larger companies, which adhere to levels of corporate governance and have lower net revenues, are more likely to disclose the measures \"EBITDA\" and \"Adjusted EBITDA\". Evidence from this study may be useful to collaborate with the current discussion of regulators and regulators by pointing out the informative role of alternative measures of measurement, while noting that these figures need to be monitored and monitored by appropriate bodies and institutions.

Diante dos problemas econômicos e sociais ocasionados pelo fenômeno da inflação nas últimas décadas, foi instituído, no Brasil, em 1994, o Plano Real, cujo objetivo era controlar os altos níveis de inflação vivenciados nas últimas décadas e alcançar a estabilidade econômica. Nesse contexto, a obrigatoriedade das práticas de correção monetária institucionalizadas foi destituída e institui-se Lei proibindo a publicação de demonstrações contábeis corrigidas. Iniciou-se, então, o processo de desinstitucionalização das práticas contábeis de correção monetária nas empresas brasileiras. Este trabalho teve como objetivo analisar o processo de desinstitucionalização das práticas contábeis de correção monetárias em empresas brasileiras. Os pressupostos teóricos utilizados nesta pesquisa foram baseados na teoria institucional que representa uma abordagem sociológica interpretativa do comportamento humano, que reconhece fenômenos de racionalidade limitada e o caráter político da ação social. Esta abordagem fundamenta-se num modelo político, no qual o comportamento social é pautado pela legitimização e pela conformidade a padrões socialmente impostos pela socialização dos indivíduos, pelos contextos das organizações e pelos mercados. Os modelos utilizados nas análises empíricas foram baseados nas abordagens propostas por Oliver (1992) e DiMaggio e Powell (1983). A estratégia de pesquisa utilizada compreendeu aplicação de questionários e entrevistas numa população de 118 empresas brasileiras, de grande porte, constantes da Revista Exame 500 Maiores Empresas e teve como principais respondentes os contadores e gerentes de controladoria. Foram utilizados a técnica estatística multivariada Análise Fatorial, o teste ANOVA one way e o teste Kruskal-Wallis. Constatou-se que o uso e manutenção das práticas contábeis de correção monetária, no contexto brasileiro, mantêm um relacionamento estreito como fatores isomórficos: coercitivos e normativos. Sendo que as variáveis mais relevantes foram: a) obrigatoriedade imposta pela lei; b) poder institucional das entidades reguladoras; c) regulamentação do governo; e d) recomendações das empresas consultoria e auditoria. Em relação ao grau de influência das pressões políticas, funcionais e sociais no processo de desinstitucionalização destacam: a) diminuição dos índices de inflação na economia; b) mudança das leis sobre publicação; c) declínio da exigência por parte do mercado financeiro; d) custos de manter a prática; e) surgimento de novas tecnologias contábeis; f) níveis de inflação no período (mês e ano); e g) harmonização internacional das práticas contábeis. A estratégia de realizar entrevistas como os gestores das empresas que mantiveram ou ainda mantém a prática contábil de correção monetária representou um recurso de grande utilidade no delineamento do fenômeno pesquisado. Pois possibilitou, basicamente, a triangulação dos resultados. Portanto, a explicação do processo de desinstitucionalização das práticas contábeis de correção monetária é decorrente de uma combinação distinta de fatores institucionais. Sendo que os fatores observados estão aderentes aos pressupostos da teoria institucional e estão em consonância com os resultados das pesquisas realizadas no mercado americano e no Reino Unido. / In face of the economic and social costs caused by inflation in previous decades, Plano Real was brought to being in Brazil in1994. Its aim was to control the existing high inflation rates and reach economic stability. In this scenario, the enforced use of institutionalized inflation adjusted practices was withdrawn and a law is passed to forbid the issuance of adjusted accounting statements. The process of desinstitutionalization of accounting practices then began in Brazilian enterprises. This thesis aimed at analyzing the desinstitutionalization of accounting practices of inflation adjustment in Brazilian enterprises. The theoretical framework used is based on the institutional theory that represents a sociological approach to interpret human behavior; it acknowledges limited rationality phenomena and the political attribute of the social action. This approach is based on a political model in which social behavior is guided by legitimization and compliance to social patterns imposed by the socialization of individuals, by the scenario of enterprises and by the markets. The models used in the empirical analyses were founded on the approaches proposed by Oliver (1992), and DiMaggio & Powell (1983). The research strategy comprised questionnaires and interviews, carried out with a population of 118 enterprises listed among the top 500 according to Revista Exame 500 Maiores Empresas, and the main respondents were accountants and controllership managers. Data Reduction Analysis, ANOVA one way test, and Kruskal-Wallis test were used. Analysis shows that, as for the Brazilian scenario, the use and continuance of inflation-adjusted practices bear close relationship with isomorphic institutional factors: coercive and regulatory. The most relevant variables were: (i) law enforcement; (ii) institutional power of regulatory organizations; (iii) governmental regulation; and (d) advice by consulting and auditing offices. Regarding the level of influence of political, functional and social pressure for the desinstitutionalization, we can emphasize: (a) the decrease of inflation rates in the economy; (b) the changes in laws on disclosure; (c) lack of demand by financial markets; (d) costs to keep the practice; (e) the emergence of new accounting technologies; (f) levels of inflation in the period (month/year), and (g) international harmonization of accounting practices. The strategy of interviewing managers of enterprises that performed or still perform inflation-adjusted practices proved a very useful tool in the characterization of the phenomenon researched. It made the triangulation of results possible. Thus, the explanation of the process of desinstitutionalization of inflation adjustment practices results from a distinct combination of institutional aspects. These adhere to the assumptions of the institutional theory and are in agreement with the results of research carried out in the markets in The USA and The United Kingdom.

Incorporating survey weights into likelihood-based analysis is a controversial issue because the sampling weights are not simply equal to the reciprocal of selection probabilities but they are adjusted for various characteristics such as age, race, etc. Some adjustments are based on nonresponses as well. This adjustment is accomplished using a combination of probability calculations. When we build a logistic regression model to predict categorical outcomes with survey data, the sampling weights should be considered if the sampling design does not give each individual an equal chance of being selected in the sample. We rescale these weights to sum to an equivalent sample size because the variance is too small with the original weights. These new weights are called the adjusted weights. The old method is to apply quasi-likelihood maximization to make estimation with the adjusted weights. We develop a new method based on the correct likelihood for logistic regression to include the adjusted weights. In the new method, the adjusted weights are further used to adjust for both covariates and intercepts. We explore the differences and similarities between the quasi-likelihood and the correct likelihood methods. We use both binary logistic regression model and multinomial logistic regression model to estimate parameters and apply the methods to body mass index data from the Third National Health and Nutrition Examination Survey. The results show some similarities and differences between the old and new methods in parameter estimates, standard errors and statistical p-values.

Objective To evaluate the cost-effectiveness of introducing the 10-valent pneumococcal conjugate vaccine (PCV10) versus the 13-valent PCV (PCV13) to the National Immunization Schedule in Peru for prevention of pneumococcal disease (PD) in children <5 years of age. Methods The integrated TRIVAC vaccine cost-effectiveness model from the Pan American Health Organization's ProVac Initiative (version 2.0) was applied from the perspective of the Government of Peru. Twenty successive cohorts of children from birth to 5 years were evaluated. Clinical outcomes were pneumococcal pneumonia (PP), pneumococcal meningitis (PM), pneumococcal sepsis (PS) and acute otitis media from any causes (AOM). Measures included prevention of cases, neurological sequelae (NS), auditory sequelae (AS), deaths and disability adjusted life years (DALYs). A sensitivity analyses was also performed. Findings For the 20 cohorts, net costs with PCV10 and PCV13 were US$ 363.26 million and US$ 408.26 million, respectively. PCV10 prevented 570,273 AOM; 79,937 PP; 2217 PM; 3049 PS; 282 NS; 173 AS; and 7512 deaths. PCV13 prevented 419,815 AOM; 112,331 PN; 3116 PM; 4285 PS; 404 NS; 248 AS; and 10,386 deaths. Avoided DALYs were 226,370 with PCV10 and 313,119 with PCV13. Saved treatment costs were US$ 37.39 million with PCV10 and US$ 47.22 million with PCV13. Costs per DALY averted were US$ 1605 for PCV10, and US$ 1304 for PCV13. Sensitivity analyses showed similar results. PCV13 has an extended dominance over PCV10. Conclusion Both pneumococcal vaccines are cost effective in the Peruvian context. Although the net cost of vaccination with PCV10 is lower, PCV13 prevented more deaths, pneumococcal complications and sequelae. Costs per each prevented DALY were lower with PCV13. Thus, PCV13 would be the preferred policy; PCV10 would also be reasonable (and cost-saving relative to the status quo) if for some reason 13-valent were not feasible. / This study was presented at 9th International Symposium of Pneumococci and Pneumococcal Diseases, Hyderabad, India, March 2014, and supported by the National Council of Science, Technology and Technological Innovation of Peru (CONCYTEC) and International Clinical Epidemiology Network (INCLEN Trust) / This study was made possible through the financial support of the Instituto Nacional de Salud (National Institute of Health, Lima, Peru) and the PROVAC Initiative of the Pan American Health Organization (Washington, DC, USA).

Objetivo: Apresentar e discutir métodos para estimar prevalências ajustadas em pesquisas clínicas e epidemiológicas, bem como desenvolver rotinas computacionais em SAS e R. Métodos: No contexto de estudo transversal, foi simulada uma amostra de 2.000 observações independentes, considerando o desfecho dicotômico diabetes, sexo como a variável de exposição e idade como variável de ajuste. As estimativas de prevalências ajustadas (IC 95%) foram estimadas pelos métodos de predição condicional e marginal, utilizando as rotinas desenvolvidas em SAS e R. O método Delta foi usado para construir os intervalos de confiança. Os resultados foram comparados com aqueles do SUDAAN (SAS-Callable), Stata e a macro %ADJ_PROP (SAS). Resultados: No exemplo simulado, 68,2% são do sexo feminino e a idade média (DP) foi 57,6 (5,0) anos, sendo 54,2 (3,9) anos em homens e 59,2 (4,6) anos em mulheres. A estimativa da prevalência global do desfecho foi de 25,3% (IC 95%:23,4-27,3); sendo 13,8% (IC 95%:11,7-16,7) e 30,7% (IC 95%:28,3-33,2), respectivamente para homens e mulheres. As estimativas de prevalências ajustadas por idade, por meio do método de predição condicional, foram de 19,6% (IC 95%:16,2-23,6) para homens, e 23,6% (IC 95%:21,2-26,1) para mulheres. Pelo método de predição marginal, as estimativas foram de 22,4% (IC 95%:18,7-26,5) para homens, e 26,3% (IC 95%:24,1-28,6) para mulheres. Conclusão: A discrepância entre as estimativas não ajustadas é devida ao confundimento pela idade. Estimativas livres de confundimento podem ser obtidas por meio das prevalências ajustadas pela idade. No entanto, a estimativa pelo método de predição condicional não engloba a prevalência global. Em virtude disso, o método de predição marginal é, geralmente, mais adequado. A rotina desenvolvida na versão para R é uma alternativa aos softwares comerciais. / Objective: To present and discuss methods to estimate adjusted prevalences for clinical and epidemiological research, and develop computational routines in SAS and R. Methods: In the context of cross-sectional study, it was simulated a sample of 2,000 independent observations, considering the dichotomous outcome diabetes, sex as the exposure variable and age as an adjustment variable. Adjusted prevalences were estimated by the conditional and marginal methods, using routines developed in SAS and R. Confidence intervals were constructed using the Delta method. The results were compared with those of the SUDAAN (SAS-callable), Stata and macro %ADJ_PROP (SAS). Results: In simulated example, 68.2% are female and the mean (SD) age was 57.6 (5.00) years old, being that 54.2 (3.94) years for men and 59.2 (4.60) years in women. The estimated global prevalence of outcome was 25.3% (CI 95%: 23.4-27.3) and 13.8% (CI 95%: 11.7-16.7) and 30.7% (CI 95%: 28.3-33.2), respectively for men and women. Estimates of adjusted prevalence for age, through the conditional method, were 19.6% (CI 95%: 16.2-23.6) for men, and 23.6% (CI 95%: 21,2-26.1) for women. For marginal method, the estimates were 22.4% (CI 95%: 18.7-26.5) for men and 26.3% (CI 95%: 24.1-28.6) for women. Conclusion: The observed discrepancy in estimates by sex, unadjusted, can be attributed to confounding due to difference in age distribution between sexes. Comparable estimates (without confounding) of the prevalences can be obtained through prevalence adjusted for age. However, the estimate for the conditional method does not comprise the global prevalence. As a result, the marginal method is in general more suitable. The developed routines can be useful for estimating adjusted prevalences, particularly the R version (an alternative to commercial software).

A principal motivação desse estudo é a análise da quantidade de sangue recebido em transfusão (padronizada pelo peso) por crianças com problemas hepáticos. Essa quantidade apresenta distribuição assimétrica, além de valores iguais a zero para as crianças que não receberam transfusão. Os modelos lineares generalizados, usuais para variáveis positivas, não permitem a inclusão de zeros. Para os dados positivos, foram ajustados tais modelos com distribuição gama e normal inversa. Também foi considerado o modelo log-normal. A análise de resíduos padronizados indicou heterocedasticidade, e portanto a variabilidade extra foi modelada utilizando a classe de modelos GAMLSS. A terceira abordagem consiste em modelos baseados na mistura de zeros e distribuições para valores positivos, incluídos recentemente na família dos modelos GAMLSS. Estes aliam a distribuição assimétrica para os dados positivos e a probabilidade da ocorrência de zeros. Na análise dos dados de transfusão, observa-se que a distribuição normal inversa apresentou melhor ajuste por acomodar dados com forte assimetria em relação às demais distribuições consideradas. Foram significativos os efeitos das variáveis explicativas Kasai (ocorrência de operação prévia) e PELD (nível de uma medida da gravidade do paciente com 4 níveis) assim como os efeitos de interação sobre a média e variabilidade da quantidade de sangue recebida. A possibilidade de acrescentar efeitos de variáveis explicativas para modelar o parâmetro de dispersão, permite que a variabilidade extra, além de sua dependência da média, seja melhor explicada e melhore o ajuste do modelo. A probabilidade de não receber transfusão depende de modo significativo somente de PELD. A proposta de um só modelo que alia a presença de zeros e diversas distribuições assimétricas facilita o ajuste dos dados e a análise de resíduos. Seus resultados são equivalentes à abordagem em que a ocorrência ou não de transfusão é analisada por meio de modelo logístico independente da modelagem dos dados positivos com distribuições assimétricas. / The main motivation of this study is to analyze the amount of blood transfusions received (by weight) of children with liver problems. This amount shows asymmetric distribution as well as present zero values for children who did not receive transfusions. The usual generalized linear models for positive variables do not allow the inclusion of zeros. For positive data, such models with gamma and inverse normal distributions were fitted in this study. Log-normal distribution was also considered. Analysis of the standardized residuals indicated heterocedasticity and therefore the extra variability was modelled using GAMLSS. The third approach consists of models based on a mixture of zeros and distributions for positive values, also recently included in the family of GAMLSS models. These models combine the asymmetric distribution of positive data and the probability of occurrence of zeros. In the data analysis of transfusion, the inverse normal distribution showed better goodness of fit to allow adjustment of data with greater asymmetry than the other distributions considered. The effects of the explanatory variables Kasai (occurrence of previous operation) and PELD level (a measure of the severity of the patient with 4 levels) and interaction effects on the mean and variability of the amount of blood received were signicant. The inclusion of explanatory variables to model the dispersion parameter, allows to model the extra variability, beyond its dependence on the average, and improves fitness of the model. The probability of not receiving transfusion depends signicantly only PELD. The proposal of a unified model that combines the presence of zeros and several asymmetric distributions greatly facilitates the fitness of the model and the evaluation of fitness. An advantage is the equivalence between this model and a separate logistic model to for the probability of the occurrence of transfusion and a model for the positive skewed data.

Visceral leishmaniasis (VL) is a serious neglected tropical disease that is endemic in 98 countries and presents a significant public health risk. The epidemiology of VL is complex. In the Americas, it is a zoonotic disease that is caused by a parasite and transmitted among humans and dogs through the bite of an infected sand fly vector. The infection also can be transmitted vertically from mother to child during pregnancy. Infected individuals can be classified as asymptomatic or symptomatic; both classes can transmit infection. In part due to its complexity, VL transmission dynamics are not fully understood. Stochastic compartmental epidemic models are a powerful set of tools that can be used to study these transmission dynamics. Past compartmental models for VL have been developed in a deterministic framework to accommodate complexity while remaining computationally tractable. In this work, we propose stochastic compartmental models for VL, which are simpler than their deterministic counterparts, but also have several advantages. Notably, this framework allows us to: (1) define a probability of infection transmission between two individuals, (2) obtain both parameter estimates and corresponding uncertainty measures, and (3) employ formal model comparisons. In this dissertation, we develop both population level and individual level Bayesian compartmental models to study both vector and vertical VL transmission dynamics. As part of this model development, we introduce a compartmental model that allows for two infectious classes. We also derive source specific reproductive numbers to quantify the contributions of different species and infectious classes to maintaining infection in a population. Finally, we propose a formal model comparison method for Bayesian models with high-dimensional discrete parameter spaces. These models, reproductive numbers, and model comparison method are explored in the context of simulations and real VL data from Brazil and the United States.