Dados de tempo de falha são caracterizados pela presença de censuras, que são observações que não foram acompanhadas até a ocorrência de um evento de interesse. Para estudar o comportamento de dados com essa natureza, distribuições de probabilidade são utilizadas. Além disso, é comum se ter uma ou mais variáveis explicativas associadas aos tempos de falha. Dessa forma, o objetivo geral do presente trabalho é propor duas novas distribuições utilizando a função geradora de distribuições gama, no contexto de modelos de regressão em análise de sobrevivência. Essa função possui um parâmetro de forma que permite criar famílias paramétricas de distribuições que sejam flexíveis para capturar uma ampla variedade de comportamentos simétricos e assimétricos. Assim, a distribuição Weibull e a distribuição log-logística foram modificadas, dando origem a duas novas distribuições de probabilidade, denominadas de gama-Weibull e gama-log-logística, respectivamente. Consequentemente, os modelos de regressão locação-escala, de longa-duração e com efeito aleatório foram estudados, considerando as novas distribuições de probabilidade. Para cada um dos modelos propostos, foi utilizado o método da máxima verossimilhança para estimar os parâmetros e algumas medidas de diagnóstico de influência global e local foram calculadas para encontrar possíveis pontos influentes. No entanto, os resíduos foram propostos apenas para os modelos locação-escala para dados com censura à direita e para dados com censura intervalar, bem um estudo de simulação para verificar a distribuição empírica dos resíduos. Outra questão explorada é a introdução dos modelos: gama-Weibull inflacionado de zeros e gama-log-logística inflacionado de zeros, para analisar dados de produção de óleo de copaíba. Por fim, diferentes conjunto de dados foram utilizados para ilustrar a aplicação de cada um dos modelos propostos. / Failure time data are characterized by the presence of censoring, which are observations that were not followed up until the occurrence of an event of interest. To study the behavior of the data of that nature, probability distributions are used. Furthermore, it is common to have one or more explanatory variables associated to failure times. Thus, the goal of this work is given to the generating of gamma distributions function in the context of regression models in survival analysis. This function has a shape parameter that allows create parametric families of distributions that are flexible to capture a wide variety of symmetrical and asymmetrical behaviors. Therefore, through the generating of gamma distributions function, the Weibull distribution and log-logistic distribution were modified to give two new probability distributions: gamma-Weibull and gammalog-logistic. Additionally, location-scale regression models, long-term models and models with random effects were also studied, considering the new distributions. For each of the proposed models, we used the maximum likelihood method to estimate the parameters and some diagnostic measures of global and local influence were calculated for possible influential points. However, residuals have been proposed for data with right censoring and interval-censored data and a simulation study to verify the empirical distribution of the residuals. Another issue explored is the introduction of models: gamma-Weibull inflated zeros and gamma-log-logistic inflated zeros, to analyze production data copaiba oil. Finally, different data set are used to illustrate the application of each of the models.

Com o avanço tecnológico aprimorado, diferentes comportamentos do tempo de vida vem sendo estudados, e com isso é necessário a criação de novos modelos, muitas vezes mais complexos, para melhor ajuste e inferência sobre a população em estudo. A distribuição beta semi-normal generalizada é útil para modelagem de tempos de vida, e com isso propomos neste trabalho uma distribuição mais ampla chamada distribuição beta semi-normal generalizada geométrica, cuja função de risco pode assumir as formas crescente, decrescente, forma de banheira ou modal. A função densidade da nova distribuição é escrita como uma combinação linear da função densidade da distribuição beta semi-normal generalizada, sendo assim, algumas importantes propriedades da nova distribuição foram obtidas, como: momentos, assimetria, curtose, função geradora de momentos, desvios médios, função quantíl e curvas de Lorenz e de Bonferroni. Para a estimação dos parâmetros, é utilizado o método de máxima verossimilhança. Também foi proposto no trabalho, o novo modelo de regressão baseado na distribuição beta semi-normal generalizada geométrica, os quais podem ser muito úteis em análise de dados reais por serem mais flexíveis. / Due to the technological improved advances, different behaviors of the lifetime has been studied and for this reason, it is necessary to create new statistical models, many times more complex, for the better fit and inferences about the population under study. The beta generalized half-normal distribution is useful for modeling lifetime data, and in this sense, we propose, in this work, a wider distribution called the geometric beta generalized half-normal distribution in which the hazard function takes the forms increasing, decreasing, bathtub and unimodal. The density function of the new distribution can be written as a linear combination of the beta generalized half-normal densities, and thereby, some properties of the new distribution can be obtained such as the moments, skewness, kurtosis, moment generating function, mean deviations, quantile function and Lorenz and Bonferroni curves. For the estimation of the parameters, we use the maximum likelihood method considering the presence of censored data. We also propose a new regression model based on the geometric beta generalized half-normal distribution, which can be very useful in the analysis of real data due to their flexibility.

Introdução. Há evidências de que a desigualdade social pode influenciar o prognóstico do câncer quando foram comparadas as taxas de sobrevida segundo o status socioeconômico, entretanto isso ainda não foi estudado no Brasil. Objetivo. Analisar a desigualdade social na determinação da sobrevida específica em um, três e cinco anos, de pacientes diagnosticados com câncer de boca e orofaringe em São Paulo. Metodologia. Utilizou-se a coorte de base hospitalar do grupo de pesquisa Genoma Clínico do Câncer de Cabeça e Pescoço (GENCAPO), à qual foram aplicados novos critérios de exclusão. Assim, foram analisados dados demográficos, socioeconômicos, comportamentais, sintomas autorreferidos e de condições clínicas de 1.154 pacientes, com diagnóstico de malignidade nas regiões de boca (C00.3-C06) e orofaringe (C09-C10), arrolados de 2001 a 2009 e acompanhados por cinco anos. A desigualdade social foi aferida pela escolaridade e a ocupação. As probabilidades acumuladas de sobrevida específica em um, três e cinco anos foram calculadas pelo método de Kaplan-Meier e as diferenças entre as curvas de sobrevida, pelo teste de Log-rank. A regressão de Cox univariada e múltipla possibilitou a análise dos fatores associados à sobrevida - Hazard Ratio (HR) com IC95%. Resultados. As probabilidades acumuladas de sobrevida específica em um, três e cinco anos para câncer de boca e orofaringe, foram de 74,72%, 49,86 e 42,46%, respectivamente. Baixa escolaridade (HR=1,25; IC95%=1,03-1,51) e realização de trabalhos manuais (HR=1,37; IC95%=1,05-1,78) foram associados à menor probabilidade de sobrevivência em cinco anos na análise não ajustada. A cor da pele preta/parda, tumor de tamanho T3 ou T4 e comprometimento de linfonodos foram fatores independentemente associados à menor sobrevida específica por câncer de boca e orofaringe em um, três e cinco anos. Disfagia e dificuldade respiratória foram associados à menor sobrevida específica em um ano. Em três anos, observou-se também a otalgia e em cinco anos, a dificuldade de deglutição. Conclusões. Foi identificada associação entre a desigualdade social, menor escolaridade e trabalho manual, à menor sobrevida em 5 anos. Esses fatores, na análise ajustada pelas características demográficas, tiveram sua associação potencialmente explicada pela cor da pele preta/parda. / Introduction. There is evidence that social inequality may influence the prognosis of cancer when comparing survival rates according to socioeconomic status, but this has not been studied in Brazil. Aim. To evaluate social inequality in the determination of one, three and five years specific survival of patients diagnosed with oral and oropharyngeal cancer in São Paulo. Methodology. A hospital-based cohort of the research group \"Clinical Genome of Head and Neck Cancer\" (GENCAPO) was used, applying new exclusion criteria. Demographic, socioeconomic, behavioral, selfreported symptoms and clinical conditions characteristics of 1,154 patients, diagnosed with malignancy in the mouth (C00.3-C06) and oropharynx (C09-C10) regions, included from 2001 to 2009 and followed up for five years, were analyzed. Social inequality was measured by schooling and occupation. The cumulative probabilities of one, three, and five years specific survival were calculated by the Kaplan-Meier method and differences between the survival curves by the log-rank test. Univariate and Multiple Cox Regression allowed to analyze the factors associated with survival - Hazard Ratio (HR) with 95% CI. Results. The accumulated probabilities of one, three and five years specific survival for oral and oropharyngeal cancer were 74.72%, 49.86 and 42.46%, respectively. Low schooling (HR = 1.25, 95% CI = 1.03-1.51) and manual work (HR = 1.37, 95% CI = 1.05-1.78) were associated with a lower survival probability at five years in the unadjusted analysis. Dysphagia and breathing difficulties were associated with lower one year specific survival. At three years, it was also observed earache, and at five years, swallowing difficulty. Conclusions. An association among social inequality, lower schooling and manual labor, with the lowest survival at 5 years, was identified. These results were potentially explained by black/brown skin color in the adjusted analysis for demographic characteristics.

As researchers increasingly use web-based surveys, the ease of dropping out in the online setting is a growing issue in ensuring data quality. One theory is that dropout or attrition occurs in phases that can be generalized to phases of high dropout and phases of stable use. In order to detect these phases, several methods are explored. First, existing methods and user-specified thresholds are applied to survey data where significant changes in the dropout rate between two questions is interpreted as the start or end of a high dropout phase. Next, survey dropout is considered as a time-to-event outcome and tests within change-point hazard models are introduced. Performance of these change-point hazard models is compared. Finally, all methods are applied to survey data on patient cancer screening preferences, testing the null hypothesis of no phases of attrition (no change-points) against the alternative hypothesis that distinct attrition phases exist (at least one change-point).

事實已經證明，會計師事務所的員工流動率相當的高，雖然高流動率是可以預期的，或甚至是事務所期望的，但若離職的員工是事務所希望留住的員工，或者是因為離職的原因並不正常，則可能會給專業界帶來一些問題。影響員工流動率的因素很多，但不論就哪一個因素來探討，均不若“組織文化”之涵蓋廣泛。任一組織均有其特殊之文化，此一文化與員工之間產生互動，進而影響員工繼續留在此一組織中之意願，因此，就組織文化對員工在職期間之影響做一探討，可對會計師事務所員工流動率的問題有一較宏觀的認知。 　　本研究基於對會計專業界未來發展之關心，以三家事務所之離職與在職員工為調查對象，利用問卷進行調查研究，試圖了解各家事務所之組織文化是否有所差異以及組織文化對員工在職期間之影響。在組織文化理論方面，藉由Schein與Ott對組織文化之概念與理論做一整合性探討，以利於對組織文化之起源與發展有一較深刻之了解，並說明其與O'Reilly, Chatman, and Caldwell所發展之OCP組織文化問卷之關聯。而根據回函之統計結果，發現三家事務所均很重視工作成果及工作細節，也非常強調團隊合作，但在穩定性與創新性上則有差異。本研究並以相關分析、變異數分析、鑑別分析及適存分析等進行樣本資料之統計分析。 　　研究結果發現： 　　1.男性之在職期間可能比女性長。 　　2.已婚者之在職期間可能比未婚者長。 　　3.教育程度愈高者其在職期間愈短。 　　4.到職日年齡愈大者，其在職期間可能愈長。 　　5.具有與會計師事務所性質相似之工作經驗者，其在職期間可能愈長。 　　6.事務所組織文化較重視穩定者，其員工之在職期間比較長。而是否具有會計師資格及績效評估好壞對員工之在職期間長短則無顯著之影響。 / High turnover in the field of public accounting is evidenced in the fact that 70%-95% of new professionals hired will leave within 6 years. While high turnover rate can be expected, it becomes a problem when the wrong people leave, or leave for the wrong reasons. There are many studies focus on individual factor or factors which influence the employee turnover. All these studies are on "micro" level. On the contrary, "organizational culture" can give us a more "macro" understanding of employee turnover. 　　This study investigate the relationship between organizational culture and employee retention in three CPA firms. I introduce the organizational culture theory based on Schein's and Ott's conceptual work. According to the descriptive statistics, I found that "Outcome", "Detail" and "Team Work" are emphasized in all firms, but significant difference exists in "Stability" and "Innovation" dimensions among three firms. This study also uses correlation analysis, ANOVA, descriminant and survival analysis to analyze data. 　　Result of the Study results suggest that 　　1. Male employee retention time may be longer than female's. 　　2. Tenure of married may be longer than that of unmarried. 　　3. The higher the education level, the shorter the retention time. 　　4. The greater the employee's age, the longer the retention time. 　　5. Employee who has prior working experience in accounting firm might stay longer than that who hasn't. 　　6. Employee retention in CPA firm which emphasizes "Stability" may be longer than that in CPA firm which emphasizes "Innovation". Female employee is more influenced by organizational culture. 　　7. Male employee that has CPA license might stay longer than that hasn't. Whether female employee had the CPA license or not may not influence her retention. 　　8. Performance evaluation doesn't have influence on employee tenure.

<p>The mortality of thoracic aortic diseases (mainly aneurysms and dissections) is high, even with surgical treatment. Epidemiology and long-term outcomes are incompletely investigated. Stroke is a major complication contributing to mortality, morbidity, and possibly to reduced quality of life. </p><p><i>Study I</i> Increasing incidence of thoracic aortic diseases 1987 – 2002 was demonstrated (n=14229). Annual number of operations increased eight-fold. Overall long-time survival was 92%, 77%, and 57% at 1, 5, and 10 years. Risk of operative and long-term mortality was reduced across time.</p><p><i>Study II</i> 2634 patients operated on the proximal thoracic aorta (Swedish Heart Surgery register) were examined. Aortic valve replacement, coronary revascularization, emergency operation, and age were independently associated with surgical death. Long-term mortality was similar for aneurysms and dissections. Operative mortality was reduced (13.7% vs 7.2%) for aneurysms but remained unchanged (22.3% vs 22.4%) for dissections across time.</p><p><i>Study III</i> 65 patients underwent selective antegrade cerebral perfusion (SACP) uni- or bilaterally. Stroke was significantly more common after unilateral SACP (29% vs 8%, p=0.045), confirmed by propensity score-matched analysis. Subclavian artery cannulation with Seldinger-technique entailed vascular complication in one case (1.5%).</p><p><i>Study IV</i> Near-infrared spectroscopy (NIRS) was used to monitor cerebral tissue saturation (rSO2) during SACP in 46 patients. Lower rSO2 were encountered (1) in patients suffering a stroke (2) with unilateral SACP, and (3) in the affected hemisphere of stroke victims. A decrease of rSO2 by 14 – 21% from baseline increased the risk of stroke significantly.</p><p><i>Study V</i> Quality of life (QoL) in 76 survivors of thoracic aortic surgery was examined with the SF-36 health questionnaire. Except for pain, QoL was reduced in all dimensions. QoL was not affected by acuity of operation. Tendencies of lower QoL after descending aortic operations, after major complications, and with persistent dysfunction were non-significant.</p>

In this thesis, we consider the problem of estimating the regression function in location-scale regression models. This model assumes that the random vector (X,Y) satisfies Y = m(X) + s(X)e, where m(.) is an unknown location function (e.g. conditional mean, median, truncated mean,...), s(.) is an unknown scale function, and e is independent of X. The response Y is subject to random right censoring, and the covariate X is completely observed. In the first part of the thesis, we assume that m(x) = E(Y|X=x) follows a polynomial model. A new estimation procedure for the unknown regression parameters is proposed, which extends the classical least squares procedure to censored data. The proposed method is inspired by the method of Buckley and James (1979), but is, unlike the latter method, a non-iterative procedure due to nonparametric preliminary estimation. The asymptotic normality of the estimators is established. Simulations are carried out for both methods and they show that the proposed estimators have usually smaller variance and smaller mean squared error than the Buckley-James estimators. For the second part, suppose that m(.)=E(Y|.) belongs to some parametric class of regression functions. A new estimation procedure for the true, unknown vector of parameters is proposed, that extends the classical least squares procedure for nonlinear regression to the case where the response is subject to censoring. The proposed technique uses new `synthetic' data points that are constructed by using a nonparametric relation between Y and X. The consistency and asymptotic normality of the proposed estimator are established, and the estimator is compared via simulations with an estimator proposed by Stute in 1999. In the third part, we study the nonparametric estimation of the regression function m(.). It is well known that the completely nonparametric estimator of the conditional distribution F(.|x) of Y given X=x suffers from inconsistency problems in the right tail (Beran, 1981), and hence the location function m(x) cannot be estimated consistently in a completely nonparametric way, whenever m(x) involves the right tail of F(.|x) (like e.g. for the conditional mean). We propose two alternative estimators of m(x), that do not share the above inconsistency problems. The idea is to make use of the assumed location-scale model, in order to improve the estimation of F(.|x), especially in the right tail. We obtain the asymptotic properties of the two proposed estimators of m(x). Simulations show that the proposed estimators outperform the completely nonparametric estimator in many cases.

Selection of signi?cant genes via expression patterns is important in a microarray problem. Owing to small sample size and large number of variables (genes), the selection process can be unstable. This research proposes a hierarchical Bayesian model for gene (variable) selection. We employ latent variables in a regression setting and use a Bayesian mixture prior to perform the variable selection. Due to the binary nature of the data, the posterior distributions of the parameters are not in explicit form, and we need to use a combination of truncated sampling and Markov Chain Monte Carlo (MCMC) based computation techniques to simulate the posterior distributions. The Bayesian model is ?exible enough to identify the signi?cant genes as well as to perform future predictions. The method is applied to cancer classi?cation via cDNA microarrays. In particular, the genes BRCA1 and BRCA2 are associated with a hereditary disposition to breast cancer, and the method is used to identify the set of signi?cant genes to classify BRCA1 and others. Microarray data can also be applied to survival models. We address the issue of how to reduce the dimension in building model by selecting signi?cant genes as well as assessing the estimated survival curves. Additionally, we consider the wellknown Weibull regression and semiparametric proportional hazards (PH) models for survival analysis. With microarray data, we need to consider the case where the number of covariates p exceeds the number of samples n. Speci?cally, for a given vector of response values, which are times to event (death or censored times) and p gene expressions (covariates), we address the issue of how to reduce the dimension by selecting the responsible genes, which are controlling the survival time. This approach enables us to estimate the survival curve when n << p. In our approach, rather than ?xing the number of selected genes, we will assign a prior distribution to this number. The approach creates additional ?exibility by allowing the imposition of constraints, such as bounding the dimension via a prior, which in e?ect works as a penalty. To implement our methodology, we use a Markov Chain Monte Carlo (MCMC) method. We demonstrate the use of the methodology with (a) di?use large B??cell lymphoma (DLBCL) complementary DNA (cDNA) data and (b) Breast Carcinoma data. Lastly, we propose a mixture of Dirichlet process models using discrete wavelet transform for a curve clustering. In order to characterize these time??course gene expresssions, we consider them as trajectory functions of time and gene??speci?c parameters and obtain their wavelet coe?cients by a discrete wavelet transform. We then build cluster curves using a mixture of Dirichlet process priors.

In survival analysis interest often lies in the relationship between the survival function and a certain number of covariates. It usually happens that for some individuals we cannot observe the event of interest, due to the presence of right censoring and/or left truncation. A typical example is given by a retrospective medical study, in which one is interested in the time interval between birth and death due to a certain disease. Patients who die of the disease at early age will rarely have entered the study before death and are therefore left truncated. On the other hand, for patients who are alive at the end of the study, only a lower bound of the true survival time is known and these patients are hence right censored. In the case of censored and/or truncated responses, lots of models exist in the literature that describe the relationship between the survival function and the covariates (proportional hazards model or Cox model, log-logistic model, accelerated failure time model, additive risks model, etc.). In these models, the regression coefficients are usually supposed to be constant over time. In practice, the structure of the data might however be more complex, and it might therefore be better to consider coefficients that can vary over time. In the previous examples, certain covariates (e.g. age at diagnosis, type of surgery, extension of tumor, etc.) can have a relatively high impact on early age survival, but a lower influence at higher age. This motivated a number of authors to extend the Cox model to allow for time-dependent coefficients or consider other type of time-dependent coefficients models like the additive hazards model. In practice it is of great use to have at hand a method to check the validity of the above mentioned models. First we consider a very general model, which includes as special cases the above mentioned models (Cox model, additive model, log-logistic model, linear transformation models, etc.) with time-dependent coefficients and study the parameter estimation by means of a least squares approach. The response is allowed to be subject to right censoring and/or left truncation. Secondly we propose an omnibus goodness-of-fit test that will test if the general time-dependent model considered above fits the data. A bootstrap version, to approximate the critical values of the test is also proposed. In this dissertation, for each proposed method, the finite sample performance is evaluated in a simulation study and then applied to a real data set.

<p>This study extends the existing literature on how social networks enhance the performance and sustainability of small enterprises. More specifically, the study isolates and investigates the mechanisms through which social capital helps with the growth and survival of MSMEs. The evidence presented in this study strongly suggests that an indigenous social network widely practiced in Ethiopia, the &ldquo / iqqub&rdquo / , contributes significantly to the start-up, survival and development of urban MSMEs.</p>