This study investigates the reasons for apparent differences in entrepreneurship rates in rural and urban areas using a Survey of Rural Kentucky Residents (SRKR) and the Panel Study of Entrepreneurial Dynamics (PSED) data. We estimate the determinants of dissimilar characteristics for rural and urban areas in two aspects: one is individual and contextual resources; the other is cultural tendencies of resources. The results of the analysis suggest that the difference in available individual, economic, and social support resources does not explain the observed difference in entrepreneurship rate. The results also indicate that gender, ethnicity, income, and number of children in the family have different effects on entrepreneurial intentions in rural and urban settings. The results suggest that policy makers need to account for cultural or geographical differences when designing entrepreneurial educational and support programs in order to enhance the establishment of new business between rural and urban areas.

Submitted by Alison Vanceto (alison-vanceto@hotmail.com) on 2017-03-07T12:28:13Z No. of bitstreams: 1 TeseDCTG.pdf: 2877933 bytes, checksum: b26c7e73a0952568aa117e1724d1bffa (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-03-20T18:57:09Z (GMT) No. of bitstreams: 1 TeseDCTG.pdf: 2877933 bytes, checksum: b26c7e73a0952568aa117e1724d1bffa (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-03-20T18:57:18Z (GMT) No. of bitstreams: 1 TeseDCTG.pdf: 2877933 bytes, checksum: b26c7e73a0952568aa117e1724d1bffa (MD5) / Made available in DSpace on 2017-03-20T19:10:02Z (GMT). No. of bitstreams: 1 TeseDCTG.pdf: 2877933 bytes, checksum: b26c7e73a0952568aa117e1724d1bffa (MD5) Previous issue date: 2016-12-05 / Não recebi financiamento / Initially, we use the quadratic transmutation maps to compose a new probability model: the transmuted log-logistic distribution. Transmutation maps are a convenient way of constructing new distributions, in particular survival ones. It comprises the functional composition of the cumulative distribution function of one distribution with the inverse cumulative distribution (quantil) function of another. Its comprehensive description of properties, such as moments, quantiles, order statistics etc., along with its survival study and the classical and Bayesian estimation methods, are also part of this work. Focusing on analysis of survival, the study included two practical situations commonly found: the presence of regression variables, through the transmuted log-logistic regression model, and the presence of right censorship. In a second moment, searching for a more exible model than the transmuted, we present its generalization, the transmuted distributions of cubic rank. Using the methodology presented in this rst generalization, two models were considered to compose the new cubic transmuted distributions: the log-logistic and Weibull models. Faced with problems presented in the transmutated class of quadratic and cubic orders (such as the restricted parametric space of the transmutation parameter ), we propose in this work, a new family of distribution. This family, which we call e-transmuted or e-extended, is as simple as the transmuted model, because it includes a single parameter to the base model, but more exible than the class of transmuted models, once the transmuted is a particular case of the proposed family. In addition, the nem family presents important properties such as, orthogonality between the baseline model parameters and the e-transmutation parameter, along with unrestricted parametric space for the ! e-transmutation parameter, which is de ned on the real line. Simulation studies and real data applications were performed for all proposed models and generalizations. / Inicialmente, usamos os mapas de transmutação quadráticos para compor um novo modelo de probabilidade: a distribuição log-logística transmutada. Mapas de transmutação são uma forma conveniente de construção de novas distribuições, em especial de sobrevivência/con abilidade, e compreendem a composição funcional da função de distribuição acumulada e da função de distribuição acumulada inversa (quantil) de um outro modelo. Uma descrição detalhada de suas propriedades, tais como, momentos, quantis, estatística de ordem, dentre outras estatísticas, juntamente com o estudo de sobrevivência e métodos de estimação clássico e Bayesiano, também fazem parte deste trabalho. Focando em análise sobrevivência, incluímos no estudo duas situações práticas comumente encontradas: a presença de variáveis regressoras, através do modelo de regressão transmutado log-logístico, e a presença de censura à direita. Em um segundo momento, buscando um modelo mais exível que o transmutado, apresentamos uma generalização para esta classe de modelos, as distribuições transmutadas de rank cúbico. Usando a metodologia apresentada nesta primeira generalização, dois modelos foram considerados para compor as novas distribuições transmutadas cúbica: os modelos log-logístico e Weibull. Diante de problemas apresentados na classe transmutada de ordens quadrática e cúbica (tal como o espaço paramétrico restrito do parâmetro de transmuta ção ), propomos neste trabalho, uma nova família de distribuição. Esta família, a qual chamamos e-transmutada ou e-extendida, é tão simples quanto o modelo transmutado, por incluir um único parâmetro ao modelo base, porém mais exível do que a classe de modelos transmutados, sendo esta classe um caso particular da família proposta. Além disso, apresenta propriedades importantes, como ortogonalidade entre os parâmetros do modelo base e o parâmetro de e-transmutação, e espaço paramétrico não restrito para o parâmetro de etransmuta ção !, que é de nido em toda reta real. Estudos de simulação e aplicações a dados reais foram realizados para todos os modelos e generalizações propostas.

nÃo hà / Tendo em vista os reconhecidos benefÃcios sociais e econÃmicos advindos da educaÃÃo, tanto em nÃvel individual quanto agregado, identificar os fatores que causam os problemas de evasÃo e repetÃncia escolar tem persistido como ponto importante na agenda de pesquisadores, seja de paÃses em desenvolvimento ou desenvolvidos. Nesse sentido, este estudo objetivou identificar os determinantes da evasÃo e da repetÃncia escolar no ensino mÃdio das escolas pÃblicas do CearÃ. Para tanto, foi utilizada a base de dados longitudinal do Sistema de AvaliaÃÃo Permanente da EducaÃÃo BÃsica (SPAECE) de 2008, acompanhada de 2009 a 2011, sobre a qual se construiu um modelo logÃstico multinÃvel que incorpora a natureza hierÃrquica dos dados, subdivididos em nÃveis de alunos e de escolas. Os resultados indicam a importÃncia do Programa Bolsa FamÃlia para a reduÃÃo da probabilidade, tanto para a ocorrÃncia de evasÃo, quanto de repetÃncia escolar. AlÃm disso, dentre os fatores que tÃm a maior influÃncia no aumento das chances de o aluno deixar a escola, estÃo o desinteresse pelos estudos e a persistÃncia de repetÃncia em anos anteriores. Sendo este Ãltimo, junto com o atraso escolar, os principais determinantes da retenÃÃo dos alunos cearenses. Uma vez que essas escolas possuem infraestruturas semelhantes, a intervenÃÃo dos educadores deve estar atrelada à motivaÃÃo dos alunos pelos estudos, atravÃs de medidas pedagÃgicas que os envolvam e os alertem efetivamente sobre os benefÃcios sociais e econÃmicos advindos pela acumulaÃÃo de educaÃÃo a mÃdio e longo prazo, tal como ocorre nos paÃses desenvolvidos. / Considering the overall benefits from education, the dropouts and school failures bring about severe consequences for economic progress and welfare of the population. The purpose of this study was to identify the determinants for the occurrence of such problems in public high schools of the State of CearÃ. For doing this, a longitudinal database of SPAECE of 2008, extended to 2009-2011, is applied to a multilevel logistic model on a hierarchical basis, which is divided in two dimensions: school and students. It is found evidence that factors associated with both the students themselves and their familiesâ status affects significantly the chances for dropouts and school failures. While the cash transfer (Bolsa FamÃlia) to studentâs household is an important mechanism to reduce the probability of dropout and school failure, the lack of students' dedications in school activities and persistency of failures are the main deficiencies for increasing the chances of dropouts from school. Since the sampled schools have similar infrastructure, the intervention of educators must be driven to student motivation in scholar activities, through educational measures that call for their attention and warn them effectively on the social and economic benefits arising from the accumulation of education in the medium and long term, as it occurs in developed countries.

Alguns aspectos devem ser considerados nos trabalhos com frutíferas como o ponto adequado de colheita e estratégias de conservação, sendo fundamental a análise do crescimento e desenvolvimento dos frutos. Vários tipos de modelos estatísticos podem descrever os processos de crescimento envolvidos no sistema de produção vegetal, sendo os modelos não lineares considerados mais adequados para esse tipo de estudo. Para que os resultados obtidos no ajuste do modelo sejam válidos é necessário verificar a qualidade de ajuste por meio de uma análise dos resíduos. A distribuição dos resíduos ordinários para os modelos de regressão não linear, é matematicamente trabalhosa e os critérios de diagnóstico são falhos, principalmente em pequenas amostras. Diante de tal fato, Cook e Tsai (1985) definiram o resíduo projetado. O comportamento dos resíduos projetados é melhor do que o dos resíduos ordinários, pois suas propriedades são mais próximas das correspondentes ao resíduo ordinário da regressão normal linear. O ganho é substancial se o teste para a medida de não linearidade for significativo. Por meio dos ajustes dos modelos logístico e de Gompertz aos dados de comprimento de peras foi realizada a análise dos resíduos ordinários e dos resíduos projetados, a fim de evidenciar suas vantagens. Os resíduos projetados se mostraram viáveis para diagnóstico dos modelos e descartaram possíveis tendências apresentadas pelos resíduos normalizados. / Some aspects must be considered in the works with fruit tree such as the appropriate point of harvest and conservation strategies, being fundamental the analysis of growth and fruit developments. Several types of statistical models can describe the growth processes involved in plant production system and the non-linear models are considered more suitable for this type of study. For the results obtained from the model fit to be valid is necessary to check the quality of fit through a residual analysis. The distribution of the ordinary residuals in the non-linear regression models is mathematically complicated and the diagnostic criteria are not precise, mainly in small samples. To overcome this fact, Cook and Tsai (1985) defined the projected residual. The behaviour of the projected residuals is better than the ordinary residuals since their properties are closer to those corresponding to the ordinary residuals for the normal linear regression. The gain is substantial if the test for non-linearity is significant. Through the fit of the logistic and Gompertz models to the pears length data, it was performed an analysis of the ordinary and projected residuals in order to demonstrate their advantages. The projected residuals showed more feasible for the diagnostic of the models and ruled out possible trends presented by normalized residuals.

O objetivo deste trabalho é apresentar os conceitos relacionados a Teste Adaptativo Informatizado, ou abreviadamente TAI, para o modelo logístico unidimensional da Teoria de Resposta ao Item. Utilizamos a abordagem bayesiana para a estimação do parâmetro de interesse, chamado de traço latente ou habilidade. Apresentamos os principais algoritmos de seleção de itens em TAI e realizamos estudos de simulação para comparar o desempenho deles. Para comparação, usamos aproximações numéricas para o Erro Quadrático Médio e para o Vício e também calculamos o tempo médio para o TAI selecionar um item. Além disso, apresentamos como instalar e usar a implementação de TAI desenvolvida neste projeto chamada de TAI2U, que foi desenvolvido no VBA-Excel usando uma interface com o R / The main of this work is to introduce the subjects related to Computerized Adaptive Testing, or breafly CAT, for the unidimensional three-parameter logistic model of Item Response Theory. We use bayesian approach to estimate the parameter of interest. We present several item selection algorithms and we perform simulations comparing them. The comparisons are made in terms of the mean square error, bias of the trait estimates, the average time for item selection and the average length of test. Furthermore, we show how to install e use the CAT implementation of this work called built in MIcrosoft Excel - VBA using interface with the statistical package R

Presented diploma thesis deals with the application of dispersion modelling of emissions from stationary sources of pollution as a potential tool which can be used for data preparation in reverse logistic models. In this case, the stationary source is represented by a waste-to-energy plant. The thesis consists of two main parts -- theoretical and computational. Firstly, emission sources and pollutants originating in thermal waste treatment are discussed. In the second chapter, legislation linked to the problem of air pollution is listed and the topics of current air quality in the Czech Republic and dispersion studies are outlined. Closing section of the theoretical part provides information about the principals of dispersion calculations and classification of dispersion models is given. In the computational part, the used dispersion software SYMOS'97 is introduced and the series of conducted calculations is presented. The calculations investigated the influence of plant capacity, dispersion conditions and terrain on the results. The input data of plant capacity and wind rose were indicated as key parameters in the calculation. The final evaluation revealed a good potential of the used dispersion model as a support tool in reverse logistic models.

The diploma thesis consists of specification of e-commerce as a tool serving for grocery presentation and distribution. Specifies conditions of foundation, operation and service relevant to used solution. It contains new framework suggestions contributing increases in effectivity and productivity in online grocery shopping.

This exploratory study assessed attitudes and perceptions of smallholder farmers towards agricultural technologies in Kakamega County, Kenya. Through a mixed-methods sequential design, the study evaluated the key variables predicting farmer adoption of agricultural innovations. While social sciences provide a clear human-driven pattern explaining the process of choices and behaviors regarding technology use, there is still little clarity on the influences of adoption decisions among smallholder farmers in rural Kenya. Using the diffusion of innovations theory, the study explored the attitudes and perceptions of smallholder farmers toward technology adoption in seven sub-counties of Kakamega County (Lurambi, Ikolomani, Shinyalu, Mumias East (Shianda), Malava Butere, and Khwisero). The study design utilized a quantitative survey of 245 smallholder heads of households, followed by focus group discussions to further probe attitudes, values and practices that could influence technology adoption. The survey questionnaire tested two hypotheses: (H1) socio-demographic characteristics are related to agricultural technology adoption; and, (H2) farmer access to extension services was related to agricultural technology adoption. A binary logistic regression model was used to quantitatively estimate socio-demographic variables presumed to influence the adoption of agricultural innovations. Subsequently, four informal focus group discussions of 28 discussants was conducted across representative sub-counties (Lurambi, Shianda, Malava and Ikolomani), to elicit an in-depth understanding of farmers’ perspectives on technology adoption. The focus group participants included farmers recruited from among survey participants. The qualitative research instrument sought to answer three questions, (RQ1) what are farmer attitudes and perceptions towards agricultural technologies; (RQ2) what socio-cultural values influence farmers’ choice of agricultural technologies; and, (RQ3) what sources do farmers use for obtaining information on agricultural technology? Quantitative results included a principal component analysis (PCA) in which 14 attitudes questions were reduced to five conceptual clusters. These clusters included: challenges in accessing modern agricultural technologies (explained 19.09% of the total variance); effectiveness of agricultural technologies (11.88%); enjoyment of agricultural technologies (10.02%); social influence in use of technology (9.47%); and experience with agricultural technologies (8.13%). A logistic regression model indicated that independently age (.07), education (.10), and off-farm income (.08) were significantly associated with adoption of technology at the 90% confidence level when controlling for all other variables in the model. However, agricultural extension (.42) was not a significant predictor of agricultural technology adoption in this model. Qualitative results provided rich insights which enhanced findings from the survey data. Key insights in the thematic analysis included: farmers’ ambivalence about agricultural technologies; lack of trust in agricultural agents; low levels of agricultural technology knowledge; extension services as the main source of information dissemination to farmers; predominance of gender in determining agricultural technology adoption; and gender inequity in agricultural decision-making. In conclusion, the study results suggested that a mixed-methods approach was valuable in probing the nuances of farmers’ perceptions of agricultural extension and technology adoption among smallholder farmers. The results supported the following recommendations: the agricultural extension efforts could be more effectively structured in order to support the dissemination of agricultural information; the issue of gender should be adequately addressed by engaging male and female in collaborative agricultural efforts to help break the barrier of gender inequity; and future research would benefit from disaggregating public and private extension services as a more robust method for determining their individual effects in the promotion of agricultural innovations among smallholder farmers.

本研究主要是希望針對採市場資訊的Merton Model、利用財務比率為變數的Logistic Model及混合用市場資訊及財務比率資訊的Hybrid Model做區別力的比較；且觀察出考慮提前違約是否能增加區別效力。進而將模型細分為: （一）以Merton Model為基礎所估出的風險中立下違約機率及客觀違約機率，及尚有為了修正Merton Model公司只會在負債到期日時才發生違約的基本假設的首次通過模型。 （二）採用Altman Z-Score所採用的五個判定公司的財務比率作為Logistic Model的基本變數。 （三）模型五、六、七除了利用Altman Z-Score所採用的五個判定公司的財務比率作為Logistic Model的基本變數。尚將風險中立下違約機率、客觀違約機率及首次通過模型違約機率分別當作的另一個解釋因子，形成同時結合市場資訊及財務資訊的Hybrid Model。 同時，採用了判別隨機性的Chi-Square Test、解釋力的Pesudo R2、區別力的S-K Test及ROC曲線、正確性的Brier Scores，此五個指標作為做為樣本內及樣本外模型的驗證方法。 經由驗證，得到的結論是同時包含市場資訊及財務資訊的模型較佳，且有考慮公司於負債到期日前就發生違約的因素對預警模型的判別效力有提高的正向作用。

In this thesis we develop several kinds of latent variable models in order to address three types of bio-statistical problem. The three problems are the treatment effect of carcinogens on tumour development, spatial interactions between plant species and motor unit number estimation (MUNE). The three types of data looked at are: highly heterogeneous longitudinal count data, quadrat counts of species on a rectangular lattice and lastly, electrophysiological data consisting of measurements of compound muscle action potential (CMAP) area and amplitude. Chapter 1 sets out the structure and the development of ideas presented in this thesis from the point of view of: model structure, model selection, and efficiency of estimation. Chapter 2 is an introduction to the relevant literature that has in influenced the development of this thesis. In Chapter 3 we use the EM algorithm for an application of an autoregressive hidden Markov model to describe longitudinal counts. The data is collected from experiments to test the effect of carcinogens on tumour growth in mice. Here we develop forward and backward recursions for calculating the likelihood and for estimation. Chapter 4 is the analysis of a similar kind of data using a more sophisticated model, incorporating random effects, but estimation this time is conducted from the Bayesian perspective. Bayesian model selection is also explored. In Chapter 5 we move to the two dimensional lattice and construct a model for describing the spatial interaction of tree types. We also compare the merits of directed and undirected graphical models for describing the hidden lattice. Chapter 6 is the application of a Bayesian hierarchical model (MUNE), where the latent variable this time is multivariate Gaussian and dependent on a covariate, the stimulus. Model selection is carried out using the Bayes Information Criterion (BIC). In Chapter 7 we approach the same problem by using the reversible jump methodology (Green, 1995) where this time we use a dual Gaussian-Binary representation of the latent data. We conclude in Chapter 8 with suggestions for the direction of new work. In this thesis, all of the estimation carried out on real data has only been performed once we have been satisfied that estimation is able to retrieve the parameters from simulated data. Keywords: Amyotrophic lateral sclerosis (ALS), carcinogens, hidden Markov models (HMM), latent variable models, longitudinal data analysis, motor unit disease (MND), partially ordered Markov models (POMMs), the pseudo auto- logistic model, reversible jump, spatial interactions.