21 |
Analys av bortfall i en uppföljningsundersökning av hälsa / Analysis of attrition in a longitudinal health studyUdd, Mattias, Pettersson, Niklas January 2008 (has links)
<p>The LSH-study started in 2003 at the department of Health and Society at the University of Linköping. The purpose of the study was to examine the relationship between life condition, stress and health. A total of 1007 people from ten different health centres in Östergötlands län participated. At the follow up, a couple of years later, 795 of the 1007 participated. 127 of the 212 in the attrition turned down the follow up, twelve people were not invited (for example in case of death) and the rest did not respond at all. The purpose of this paper is to find out in what degree the attrition in the follow up can be predicted using the information from the first survey and which variables are important. The differences between different types of attrition have also been examined. Simple and multiple bi- and multinomial logistic regression have been used in the analysis.</p><p>In total 34 variables were examined and in the final model six variables remained with a significant relation to the attrition. High BMI, regular smoking, high pulse and lack of daily exercise at the first survey were connected to a higher risk for an individual to not participate at the follow up. It is interesting that these factors are considered as risk factors for unhealthy living. Other factors related to a higher attrition were unemployment in the last year before the first survey and if the individual had parents born in another country than Sweden. The risk for attrition increased gradually when more risk factors were shown by the individual. The factors contributing an individual to turn down the follow up instead of not responding at all was if he or she were in the older age segments in the survey or if they were not active in any type of association.</p>
|
22 |
Study of Efficiency, Output Loss and Soil Erosion in Fiji's Ginger IndustryWaisiki Naqarase Gonemaituba Unknown Date (has links)
The ginger industry is one of the key industries identified by the Fiji government in its diversification strategy to accommodate the remnants of the withdrawal of the European Union’s sugar preferences. There is considerable pressure on small industries such as ginger in search of ways in which they can be made to operate efficiently and sustain the economy. Expansion of commercial agriculture into marginal land which is unsustainable adds enormous pressure on land causing soil erosion. Coupled with this is the quality issue which is a serious problem of ginger production and has reduced its competitiveness over the years. This study focuses on two types of losses in ginger production to provide an integrated approach to policy making and computing production losses. One is the observable output loss at the farm site that is not sold due to sub-standard quality related to disease and the other is the unobserved output loss due to inefficient production. The research attempts to answer the question of whether the Fiji ginger farmers are producing efficiently, and at what levels. The relative importance of each input in ginger production is examined. The study undertakes to determine the effects of key variables on farm efficiency. Also examined is the overall farm profiles based on the efficiency rankings of the ginger producers. Furthermore, this research attempts to determine factors that influence soil erosion, and those that influence the observed ginger loss. Using cross-sectional data from a ginger farm survey conducted in June 2007, this research estimates a stochastic production frontier which incorporates soil erosion as an input in the framework. Very few studies have looked at the impact of soil erosion in this context; hence, this study fills the gap by incorporating land quality in the analysis. Farms were found to produce at 69% of their maximum potential output and soil erosion resulted in 6.8% loss in ginger output. This also implies that using the same resources, technology and farming techniques efficiently can lead to a 31% increase in output. While unobserved loss to farm income is a 27% (F$4.6m) increase over the observable loss at farm site, the revenue loss to the whole industry is at least 30% (F$5.07m). Profit was a key determinant of both losses, but staying on farm, slope of land, manure use and hot water treatment affected the observed but not unobserved loss. Although farmer education had no effect on both losses, it was important for undertaking soil conservation. Fiji is in a good position to increase production as education, age and experience of farmers were not significant determinants of efficiency. Thus, displaced farmers from sugar cane farming (given serious concerns of the viability of that industry) can be encouraged to move with ease into ginger as an alternative livelihood. Lastly, the study highlights some practical implication which calls for an integrated package of policies related to use of best farming techniques, land tenure and, agricultural extension and support services for sustainable agricultural growth.
|
23 |
Uma conexão entre Binômio de Newton e ProbabilidadeCunha, Leandro Solano Carneiro da 12 April 2017 (has links)
Submitted by Marcos Samuel (msamjunior@gmail.com) on 2017-06-26T11:07:04Z
No. of bitstreams: 1
DissertacaoLeandro.pdf: 1104651 bytes, checksum: 93690896b2332038b7b798d10e81fa28 (MD5) / Approved for entry into archive by Vanessa Reis (vanessa.jamile@ufba.br) on 2017-06-29T13:01:37Z (GMT) No. of bitstreams: 1
DissertacaoLeandro.pdf: 1104651 bytes, checksum: 93690896b2332038b7b798d10e81fa28 (MD5) / Made available in DSpace on 2017-06-29T13:01:37Z (GMT). No. of bitstreams: 1
DissertacaoLeandro.pdf: 1104651 bytes, checksum: 93690896b2332038b7b798d10e81fa28 (MD5) / A Proposta deste trabalho, a princípio, é utilizar o Teorema Binomial para cálculos
de probabilidade, estabelecendo uma conexão entre esses conteúdos. A ideia é viabilizar aplicações do Teorema Binomial utilizando exemplos práticos como, por exemplo, lançamento de dados, viciados ou não, lançamento de moedas, entre outros. Será feita, também, uma extensão para o teorema multinomial, que possibilitará, através de expressões do tipo (a+b+c+...)n , determinar probabilidades quando da ocorrência de três ou mais eventos. Para tanto, deve-se ter como base conceitos referentes aos conteúdos de Combinatória e Probabilidade, que são estudados no Ensino Médio, para que os objetivos do trabalho sejam alcançados de maneira satisfatória.
|
24 |
Determinantes socioeconômicos da dinâmica de fertilidade no Nordeste brasileiro utilizando a base de dados da PCSVDF MulherCoelho, José Eduardo Holanda Ellery January 2017 (has links)
COELHO, José Eduardo Holanda Ellery. Determinantes socioeconômicos da dinâmica de fertilidade no Nordeste brasileiro utilizando a base de dados da PCSVDF mulher / José Eduardo Holanda Ellery Coelho. – 2017. 74f. Dissertação (mestrado ) – Universidade Federal do Ceará , Programa de Pós Graduação em Economia, CAEN, Fortaleza, 2017. / Submitted by Mônica Correia Aquino (monicacorreiaaquino@gmail.com) on 2017-09-21T19:36:34Z
No. of bitstreams: 1
2017_dis_jehecoelho.pdf: 9030844 bytes, checksum: abd289b8c4b6940e963eac340c375fe9 (MD5) / Approved for entry into archive by Mônica Correia Aquino (monicacorreiaaquino@gmail.com) on 2017-09-21T19:37:43Z (GMT) No. of bitstreams: 1
2017_dis_jehecoelho.pdf: 9030844 bytes, checksum: abd289b8c4b6940e963eac340c375fe9 (MD5) / Made available in DSpace on 2017-09-21T19:37:43Z (GMT). No. of bitstreams: 1
2017_dis_jehecoelho.pdf: 9030844 bytes, checksum: abd289b8c4b6940e963eac340c375fe9 (MD5)
Previous issue date: 2017 / The human fertility depends of one long list of factors, including health, nutrition, sexual
behavior, culture, endocrinology, time, economy, life style and emotions. Fertility rates
are different between the countries and cultures because these factors are variable. Due
to the influence of several elements that changes the dynamics of fertility, this paper has
as proposal to determinate the dynamics of women fertility in the Brazilian Northeast
through the Analisys of Categorical Sequences, in order to delineate existing patterns
and identify their variations. In this way, starting of the socioeconomic/demographic and
cultural variables, and informations of fertility dynamics, the Multinomial logistc model
was applied with the objective of obtainig more prepoderant variables about each situation
outlined. With the utilization of the sampled database of PCSV DF Woman, referring to
women from the capitals of the states of Brazilian Northeast. In view of this, we found 4
well-defined groups, which presented differences in relation to Race, Income, Age, Religion
and Schooling. The strong predictor of fertility dynamics is women’s level of education.
The highest levels of schooling corresponds to the lowers fertility rates. / A fertilidade humana depende de uma longa lista de fatores, incluindo saúde, nutrição,
comportamento sexual, cultura, endocrinologia, tempo, economia, modo de vida e emoções.
As taxas de fecundidade são diferentes entre países e culturas porque esses fatores são
variáveis. Dada à influência de diversos elementos que alteram a dinâmica de fertilidade,
este trabalho tem como proposta determinar as dinâmicas de fertilidade das mulheres
no Nordeste brasileiro através da Análise de Sequências Categóricas, a fim de delinear
os padrões existentes e identificar suas variações. Dessa forma, a partir das variáveis
socioeconômicas/demográficas e culturais, e a informações das dinâmicas de fertilidade,
aplicou-se o Modelo Logístico Multinomial com o objetivo de obter as variáveis mais
preponderantes acerca de cada dinâmica delineada. Com a utilização da base de dados
amostrados da PCSV DFMulher, referente às mulheres das capitais dos estados do Nordeste
brasileiro. Diante disto, constatou-se 4 grupos bem definidos, os quais apresentam diferenças
em relação a Raça, Renda, Idade, Religião e Escolaridade. O forte preditor da dinâmica
de fertilidade é o nível de escolaridade da mulher. Os níveis mais altos de escolaridade
correspondem as taxas de fecundidade mais baixas.
|
25 |
Aplicativo computacional para a análise de experimentos envolvendo variáveis respostas categorizadasFurtado, Mara Lúcia Ceschini [UNESP] 02 1900 (has links) (PDF)
Made available in DSpace on 2014-06-11T19:24:43Z (GMT). No. of bitstreams: 0
Previous issue date: 2003-02Bitstream added on 2014-06-13T20:32:04Z : No. of bitstreams: 1
furtado_mlc_me_botfca.pdf: 219106 bytes, checksum: d3f19324f29700359ca81643a4b286b6 (MD5) / Na experimentação agronômica existem várias situações em que a variável biológica é avaliada por meio de categorias de respostas (qualidades de um atributo), cuja escala de mensuração é nominal ou ordinal. Nestas situações, a utilização de procedimentos estatísticos que envolvam a distribuição multinomial de probabilidades tornase mais adequada e a interpretação dos resultados propicia a melhora da qualidade de discussão biológica do fenômeno. A literatura especializada, entretanto, tem mostrado que na área de Ciências Agronômicas, Florestais e Agropecuárias, muito pouco tem-se produzido com o uso de testes multinomiais apropriados, como por exemplo, o teste de associação de Goodman e seus recorrentes. Dessa forma, visando estudar a utilização de contrastes lineares entre e dentro de proporções multinomiais de experimentos nestas áreas, objetivou-se elaborar um programa computacional de alto nível, mas de simples manuseio e fácil acesso a todos os pesquisadores das áreas aplicadas. Pretende-se ainda anexar ao procedimento, o manual do usuário com exemplos de operacionalização do software e sua contribuição no estudo da incidência de fungos que causam lesões em plantas de milho e também no estudo da incidência de diarréia de leitões a partir do uso de silagem de grãos úmidos de milho nas rações iniciais após o desmame de leitões. / In Agronomical experimentation there are several situations in which the biological variable is evaluated through categories of answers (qualities of a predicate), whose measuring scale is nominal or ordinal. In these situations, the utilization of statistic procedures involving the multinominal distribution of probabilities becomes more adequate within the viewpoint of the use of quantitative methods, and the discussion about the test results improves the quality of the phenomenon biological interpretation. The specialized literature, however, has shown that in the field of Forestal and Agronomical Sciences, very little has been produced with the use of appropriate multinomial tests, as for example, Goodman association test and its recurrent ones. Thus, with the aim of studying the utilization of linear contrasts among and in the multinomial proportions of experiments in these areas, the objective is to elaborate a high level computing program, but of simple handling and easy access to all researchers of the applied areas. It is still intended to add to the procedure the user’s handbook with examples of software operation and its contribution of the observational findings about the incidence of fungi which cause lesions on corn plants and also to the study about incidence of diarrhea in piglets from the use of silage of humid grains of corn for initial feeds after piglet wearing.
|
26 |
Probabilidade de acesso e gastos com medicamentos no Brasil: estudos a partir de dados da PNAD e da POFda Silva Vasconcelos, Rafael 31 January 2011 (has links)
Made available in DSpace on 2014-06-12T17:21:58Z (GMT). No. of bitstreams: 2
arquivo6593_1.pdf: 1372457 bytes, checksum: 7d328b5860db2137060c9694c0610e5c (MD5)
license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5)
Previous issue date: 2011 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Essa dissertação é dividida em dois ensaios. O primeiro objetiva caracterizar a
demanda por medicamentos no Brasil. Para isso é feita uma revisão da literatura sobre
o tema e expostas suas particularidades para o caso brasileiro por meio dos microdados
da POF 2002-2003. Com isso são estimadas as probabilidades de gastos por categoria
de medicamento. Conforme os resultados, em geral, essas probabilidades são crescentes
com a renda e o grau de instrução do chefe da família. Contudo, outras características
como raça, região e idade possuem relevantes efeitos sobre essas probabilidades, onde isso
dependerá também do medicamento adquirido. O segundo ensaio visa apresentar como se dá
o acesso aos medicamentos pelos brasileiros. Para tanto são aplicados modelos de Escolha
discreta, para em seguida, com os dados da PNAD-Saúde 2009, a partir de estimações por
Multinomial Logit e Nested Logit, realizar inferências sobre a tomada de decisão de consumo
dos indivíduos. Os resultados sugerem que a renda, o grau de instrução e o tipo de doenças
infringidas são fatores relevantes. Onde a presença deste último fator afeta diretamente a
tomada de decisão de consumo dos brasileiros, principalmente, no caso das doenças mais
danosas à saúde
|
27 |
Estimating irrigation water demand with a multinomial logit selectivity modelHendricks, Nathan January 1900 (has links)
Master of Science / Department of Agricultural Economics / Jeffrey M. Peterson / Understanding irrigation water demand is vital to policy decisions concerning water scarcity. This thesis evaluates irrigation water-use responses to changes in prices, while accounting for cross-sectional characteristics of irrigators’ resource settings. An irrigator’s profit-maximizing decision is modeled in two stages. In the first stage, he decides which crop to plant, and in the second stage he decides how much water to apply given the crop choice.
This thesis employs an econometric modeling technique not previously used in the irrigation water demand literature, a multinomial logit selectivity model. This econometric technique allows the intensive (change in water use for each crop in the short run) and extensive (change in water use in the long run due to changes in crop-choice) margin effects to be computed in a simultaneous equation system. A multinomial logit selectivity model has applications to many resource issues in production agriculture where the two-stage decision process is common. The model is estimated from field-level data on water use and crop-choice for a 25-county region in western Kansas over the period 1991-2004.
Water use was found to be highly inelastic to the price of natural gas, but becomes more elastic as the price increases. The intensive margin effect was significant for natural gas price. The extensive margin effect only comprised half the total effect under high natural gas prices and was negligible for low prices. However, the extensive margin effect under high natural gas prices declined over time due to more efficient irrigation systems and improved crop varieties. The intensive margin effect explained most of the water use response from changes in other variables, including corn price. An increase in corn price has a negligible extensive margin effect because corn is most often substituted with alfalfa, which has a similar water requirement.
Inelastic demand implies that policies aiming to conserve the Ogallala Aquifer by increasing the price of water will not accomplish their purpose and will affect irrigators’ incomes. More effective policies would be voluntary or mandatory quantity restrictions. However, efficient restrictions would need to account for spatial variation in the rate of depletion and the remaining saturated thickness.
|
28 |
El efecto de la retroalimentación de los consumidores sobre la planificación del surtidoCarrasco Heine, Óscar Felipe January 2017 (has links)
Magíster en Gestión de Operaciones. Ingeniero Civil Industrial / Los principales antecedentes y la motivación del presente trabajo provienen de la teoría de
Assortment, los modelos de elección discreta y los procesos de Social Learning. Se apunta a
reunir parte de la vasta literatura en torno a estos temas, proponiendo un modelo matemático
que incorpore el feedback de los consumidores en la elección del surtido de productos a ofrecer
por parte de un vendedor.
El modelo propuesto incluye un comerciante minorista quien se enfrenta a una secuencia
finita de clientes, los que tienen la opción de escoger alguno de los productos ofrecidos por el
vendedor. En caso de comprar alguno, descubren su calidad y la reportan de manera pública,
siendo esta información utilizada por futuros consumidores en su proceso de decisión mediante
actualización bayesiana de sus creencias. Estos reportes a la vez permiten al retailer - quien
sólo puede ofrecer una cantidad limitada de productos - optimizar su oferta, escogiendo aquel
conjunto de alternativas que le entregue un mayor beneficio esperado.
La manera natural para abordar el problema de optimización de assortment enfrentado
por el vendedor es mediante Programación Dinámica Estocástica, al tratarse de una situación
en tiempo discreto donde período a período se debe tomar una decisión ante un determinado
estado del sistema. Sin embargo, el número de estados posibles aumenta explosivamente en
función de los parámetros considerados, volviéndose un problema virtualmente imposible de
resolver en instancias realistas. Es por ello que se opta por Programación Dinámica Aproximada,
proponiéndose 3 heurísticas que permiten obtener resultados en contextos en los cuales
no es posible optimizar de manera exacta.
A modo de referencia, se compara el rendimiento de las heurísticas con el de una estrategia
bajo la cual el vendedor no considera las evaluaciones hechas por los clientes, ni la incidencia
de sus decisiones en períodos futuros. La diferencia en los beneficios llega a ser enorme,
observándose en los experimentos realizados aumentos superiores a 300% al utilizar alguno
de los algoritmos sugeridos, en vez de obviar la información disponible. Esto no solo valida
la calidad de los métodos heurísticos propuestos, sino que sobre todo ilustra la importancia
de considerar el feedback de los consumidores. / Este trabajo ha sido parcialmente financiado por CONICYT
|
29 |
Specification, estimation and testing of treatment effects in multinomial outcome models : accommodating endogeneity and inter-category covarianceTang, Shichao 18 June 2018 (has links)
Indiana University-Purdue University Indianapolis (IUPUI) / In this dissertation, a potential outcomes (PO) based framework is developed for
causally interpretable treatment effect parameters in the multinomial dependent variable
regression framework. The specification of the relevant data generating process (DGP) is
also derived. This new framework simultaneously accounts for the potential endogeneity
of the treatment and loosens inter-category covariance restrictions on the multinomial
outcome model (e.g., the independence from irrelevant alternatives restriction).
Corresponding consistent estimators for the “deep parameters” of the DGP and the
treatment effect parameters are developed and implemented (in Stata). A novel approach
is proposed for assessing the inter-category covariance flexibility afforded by a particular
multinomial modeling specification [e.g. multinomial logit (MNL), multinomial probit
(MNP), and nested multinomial logit (NMNL)] in the context of our general framework.
This assessment technique can serve as a useful tool for model selection. The new
modeling/estimation approach developed in this dissertation is quite general. I focus here,
however, on the NMNL model because, among the three modeling specifications under
consideration (MNL, MNP and NMNL), it is the only one that is both computationally
feasible and is relatively unrestrictive with regard to inter-category covariance. Moreover,
as a logical starting point, I restrict my analyses to the simplest version of the model – the
trinomial (three-category) NMNL with an endogenous treatment (ET) variable conditioned
on individual-specific covariates only. To identify potential computational issues and to assess the statistical accuracy of my proposed NMNL-ET estimator and its implementation
(in Stata), I conducted a thorough simulation analysis. I found that conventional
optimization techniques are, in this context, generally fraught with convergence problems.
To overcome this, I implement a systematic line search algorithm that successfully resolves
this issue. The simulation results suggest that it is important to accommodate both
endogeneity and inter-category covariance simultaneously in model design and estimation.
As an illustration and as a basis for comparing alternative parametric specifications with
respect to ease of implementation, computational efficiency and statistical performance,
the proposed model and estimation method are used to analyze the impact of substance
abuse/dependence on the employment status using the National Epidemiologic Survey on
Alcohol and Related Conditions (NESARC) data.
|
30 |
Examining preferences for prevention of Louisiana's wetland lossMoore, Ross Gordon 01 May 2010 (has links)
This study analyzed preferences for wetland-loss prevention in coastal Louisiana. Data were obtained through a contingent-valuation mail survey of a random sample of Louisiana households. Results, based on 511 responses, indicate that respondents have a strong preference for a short-run program (72.41% chose this program over a long-run program or no action). Respondents that had higher incomes, were white, had prior knowledge of ongoing restoration efforts, and had confidence in government were more likely to support some program relative to no action, as were those citing hurricane, environmental, and/or climate-change protection as their primary concern. Older respondents and those with negative perceptions of climate change were more likely to prefer the short-run over the long-run program. Median net present value of willingness to pay (assuming 18.37% discount rate) was estimated at $17,491 per household for the multinomial logit model and $3,307 under the Turnbull lower-bound method.
|
Page generated in 0.1017 seconds