• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 120
  • 51
  • 9
  • 8
  • 7
  • 6
  • 4
  • 4
  • 3
  • 3
  • 2
  • 2
  • 2
  • 1
  • 1
  • Tagged with
  • 251
  • 251
  • 161
  • 32
  • 32
  • 31
  • 28
  • 28
  • 26
  • 23
  • 21
  • 19
  • 18
  • 17
  • 16
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
151

SEnsembles – uma abordagem para melhorar a qualidade das correspondências de instâncias disjuntas em estudos observacionais explorando características idênticas e ensembles de regressores

Borges Junior, Sergio Ricardo 16 December 2016 (has links)
Submitted by Ronildo Prado (ronisp@ufscar.br) on 2017-07-19T10:44:04Z No. of bitstreams: 1 TeseSRBJ.pdf: 5473127 bytes, checksum: 6ad9d0f7d24cadafbff18e445b8736d1 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-07-19T10:44:16Z (GMT) No. of bitstreams: 1 TeseSRBJ.pdf: 5473127 bytes, checksum: 6ad9d0f7d24cadafbff18e445b8736d1 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-07-19T10:44:27Z (GMT) No. of bitstreams: 1 TeseSRBJ.pdf: 5473127 bytes, checksum: 6ad9d0f7d24cadafbff18e445b8736d1 (MD5) / Made available in DSpace on 2017-07-19T10:45:12Z (GMT). No. of bitstreams: 1 TeseSRBJ.pdf: 5473127 bytes, checksum: 6ad9d0f7d24cadafbff18e445b8736d1 (MD5) Previous issue date: 2016-12-16 / Não recebi financiamento / Introduction. The datasets used in observational studies have instances belonging to two distinct groups (i.e. treatment group and control group), which are compared in order to estimate the effect of the treatment over the results. For such, in one of the approaches, called Propensity Score Matching (PSM), the propensity score for the instances of both groups is estimated and, subsequently, the correspondence of these instances is performed based on the values for the propensity score. The propensity score is the probability of attribution of a treatment based on the observed characteristics (e.g. income, sex and age). In this context, the logistic regression is widely used to estimate the propensity score and there is an great variety of instance correspondence methods. Objective. This doctor´s thesis has as its main objective to investigate computational alternatives in order to improve the quality of the instance correspondence in datasets that are manipulated in observational studies. Methodology. Techniques that estimate the propensity score and methods to perform the instance correspondence in observational studies were investigated. Thus, it was possible to investigate how the identical characteristics of the instances could be exploited in a new process to perform correspondence and, how ensembles could substitute the logistic regression by estimating the propensity scores of the instances, in the context of the PSM process. Proposal. This thesis proposes a new approach in the context of the PSM process, called “SEnsembles”, which aims to improve the quality of instance correspondence based on two main processes, which use techniques that separately consider the identical characteristics of the instances and the ensembles of regressors, more precisely, bagging, random forest and boosting. Results. The proposed approach “SEnsembles” improves the quality of the instance correspondence for the majority of calipers used (i.e. zero, 0.05, 0.10, 0.15, 0.20, 0.25 and 0.30) when compared to the baseline Nearest Neighbor Matching (NNM). Based on the experiments, when there was an improvement over the baseline, the technique that separates the identical characteristics of the instances presented improvements of up to 53.8% in the quality of correspondence, with an average of gains of 12.1%; and only 2.7% of average in the reduction of the number of pairs of instances matched. The technique which substituted the logistic regression for ensembles of regressors, in turn, presented the best correspondence with the caliper zero and with the values 0.20, 0.25 and 0.30, with improvements of up to 36.3% and an average of gains of 12.7%; and a slightly reduction of 7.6% in the number of pairs of instances matched. / Introdução. Os conjuntos de dados manipulados em estudos observacionais possuem instâncias pertencentes a dois grupos distintos (i.e. grupo de tratamento e grupo de controle), as quais são comparadas para estimar o efeito do tratamento sobre os resultados. Para isso, em uma das abordagens, chamada de Propensity Score Matching (PSM), estima-se o escore de propensão para as instâncias de ambos os grupos e, em seguida, efetua-se a correspondência dessas instâncias com base nos valores dos escores de propensão. O escore de propensão é a probabilidade de atribuição de um tratamento com base nas características observadas (por exemplo, renda, sexo e idade). Neste contexto, a regressão logística é amplamente utilizada para estimar o escore de propensão e há uma ampla variedade de métodos de correspondência de instâncias. Objetivo. Esta pesquisa de doutorado tem como objetivo principal investigar alternativas computacionais para melhorar a qualidade das correspondências de instâncias em conjuntos de dados que são manipulados em estudos observacionais. Metodologia. Investigou-se técnicas que estimam o escore de propensão e métodos para se efetuar a correspondência das instâncias em estudos observacionais. Assim, foi possível investigar como as características idênticas das instâncias poderiam ser exploradas em um novo processo de correspondência e, como ensembles, mais precisamente, bagging, random forest e boosting, poderiam substituir a regressão logística ao estimar os escores de propensão das instâncias, no contexto do processo de PSM. Proposta. Esta pesquisa propõe uma nova abordagem no contexto do processo PSM, denominada “SEnsembles”, que visa melhorar a qualidade da correspondência das instâncias com base em 2 processos principais, os quais utilizam técnicas que considerem em separado as características idênticas das instâncias e os ensembles de regressores, mais precisamente, bagging, random forest e boosting. Resultados. A abordagem proposta “SEnsembles” melhorou a qualidade da correspondência de instâncias para a maioria dos calipers utilizado (zero, 0,05, 0,10, 0,15, 0,20, 0,25 e 0,30) quando comparada ao baseline Nearest Neighbor Matching (NNM). Com base nos experimentos, quando houve ganho, a técnica que separa as características idênticas das instâncias proporcionou ganhos de até 53,8% na qualidade da correspondência, com média de 12,1% de melhoria e 2,7% de redução média do número de pares de instâncias correspondidas. Já a técnica que substituiu a regressão logística pelos ensembles proporcionou as melhores correspondências com o caliper zero e com os valores 0,20, 0,25 e 0,30, com ganhos de até 36,3% e, com média de 12,7% de melhoria e 7,6% de redução do número de pares de instâncias correspondidas.
152

Impactos do Programa Nacional de Alimentação Escolar PNAE sobre a nutrição dos alunos, defasagem e desempenho escolar

Maria Fonseca Pereira Oliveira Gomes, Sónia 31 January 2009 (has links)
Made available in DSpace on 2014-06-12T17:16:28Z (GMT). No. of bitstreams: 2 arquivo3796_1.pdf: 1367852 bytes, checksum: 0054b935a4cb8086de5756028f9e677a (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2009 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / A tese tem como objetivos: (i) avaliar o impacto do Programa Nacional de Alimentação Escolar (PNAE) sobre o nível nutricional; (ii) investigar o efeito da carência e do distúrbio nutricional sobre a defasagem idade-série e sobre o desempenho de escolas públicas brasileiras de 1ª a 8ª séries do ensino fundamental quando avaliados em testes de proficiência. Existe uma carência de estudos no Brasil a respeito dos efeitos da subnutrição sobre o desempenho escolar dos estudantes. Fato que é, em parte, explicado pela carência de dados que avalie de forma representativa o estado nutricional dos estudantes no Brasil. Esta tese, contudo utiliza dados de uma pesquisa recente ASBRAN com informações do perfil nutricional de aproximadamente 20.000 alunos de 1110 escolas públicas brasileiras. A amostra não só representa a população de estudantes de escolas públicas no Brasil, como também as informações levantadas viabilizam o estudo dos objetivos propostos. Completando os dados da ASBRAN são usados dados municipais do IBGE e do IPEA provenientes do Censo 2000 e dados educacionais do INEP 2007. Primeiramente, a tese investiga a relação do aluno com a merenda escolar. Constatou-se, por exemplo, que estudantes com carência nutricional apresentavam maior probabilidade de irem à escola apenas por conta da merenda. Além disso, ficou comprovado que municípios de maior vulnerabilidade apresentam maiores probabilidades de seus alunos irem à escola apenas pela alimentação escolar. A seguir, a tese investiga os efeitos do PNAE sobre o estado nutricional dos alunos das escolas públicas. Para isso, utilizou-se a técnica de Propensity Score. Verificou-se que o PNAE contribui para a melhoria dos desequilíbrios nutricionais registrados pelos alunos de escolas públicas A análise de impacto da carência nutricional (subnutrição) e do distúrbio alimentar sobre a defasagem idade-série envolveu o uso da distribuição binomial negativa pelo fato da variável dependente ser de contagem. Os resultados mostram que a carência nutricional tem impacto direto sobre a defasagem idade-série no Brasil, este efeito é estatisticamente significante a 99 por cento de nível de confiança e permanece inalterado à medida que se adiciona variáveis explicativas que descrevem as condições socioeconômicas das crianças, dos municípios e as características físicas das escolas. Os resultados mostram que os alunos que apresentam distúrbio nutricional apresentam nível menor de defasagem idade-série, contrariando os resultados obtidos por alguns autores. Por outro lado, os resultados das estimações de impacto do PNAE sobre o desempenho da escola não são conclusivos para a maioria das variáveis testadas. No que diz respeito ao desempenho da escola em testes de proficiência, não parece existir correlação entre a performance da instituição de ensino e o estado nutricional do aluno
153

[en] THE ADDED WORKER EFFECT FOR MARRIED WOMEN AND CHILDREN IN BRAZIL: A PROPENSITY SCORE APPROACH / [pt] O EFEITO TRABALHADOR ADICIONAL PARA MULHERES CASADAS E FILHOS NO BRASIL: UMA ABORDAGEM UTILIZANDO PROPENSITY SCORE

DANIEL GOMES DA SILVA 09 March 2017 (has links)
[pt] O efeito trabalhador adicional (AWE, em inglês) é o aumento na probabilidade de um indivíduo ingressar na força de trabalho em resposta a uma perda de emprego do chefe de família. Essa dissertação estima o AWE para mulheres casadas e filhos no Brasil utilizando métodos de propensity score matching. Encontramos evidência de um AWE significante para ambos os grupos, em particular para os filhos. Também investigamos algumas heterogeneidades no AWE para filhos. Nossos resultados sugerem que a magnitude do AWE para eles é maior para mulheres e para aqueles fora da escola. A magnitude desse AWE também está relacionada a outras características de membros do domicílio, especialmente os rendimentos do chefe de família e acesso ao seguro-desemprego. / [en] The added worker effect (AWE) is the increase in the likelihood of an individual to enter the labor force in response to the household head s job loss. This dissertation estimates the AWE for married women and children and young adults in Brazil using propensity score matching methods. We find evidence of a significant AWE for both groups, in particular for children and young adults.We also investigate some heterogeneities in the AWE for children and young adults. Our results suggest that the magnitude of the AWE for them is higher for females and for those out of school. This AWE magnitude is also related to other household members characteristics, specially the household head s earnings and access to unemployment insurance.
154

OUTPUT-INFLATION TRADE-OFF AFTER A QUARTER OF A CENTURY OF INFLATION TARGETING / Output-Inflation Trade-off After a Quarter of a Century of Inflation Targeting

Kamarád, Martin January 2016 (has links)
This thesis estimates the treatment effect of inflation targeting adoption on inflation, inflation variability, output, and output variability for 25 explicit inflation targeting countries. I implement the propensity score matching methodology that takes into account the problems of non-experimental nature, such as selection bias or selection on observable, and allows me to effectively mimic properties of randomized experiment and compute unbiased treatment effect estimates. I introduce a variety of propensity score matching methods that were recently developed in the treatment effect literature, including Nearest Neighbor, Radius matching, Kernel matching, and Inverse Probability Weighting. The results indicate that both industrial and developing inflation targeting countries exhibit lower inflation levels and at the same time higher output growth than non-targeting countries. The estimates are however in most cases statistically insignificant. Moreover, it appears that both industrial and developing countries achieve combination of lower inflation variability and output variability compared to non-targeting countries. Nonetheless, majority of the estimates are again statistically insignificant. The results are to a small extent sensitive to the choice of propensity score matching method. Radius matching with tight calipers (r=0.005, r=0.001) tends to provide the most reliable estimates. Balancing properties of the models are reasonable and compared to the previous research the standardised biases are quantitatively better.
155

Clinical comparative effectiveness of independent non-medical prescribers for type 2 diabetes

Abutaleb, Mohammed January 2015 (has links)
Independent and supplementary prescribing are the two main forms of non-medical prescribing (NMP) that have been practised in the UK since 2006. Most available studies have qualitatively investigated the impact of NMP, especially in primary care. This may be due to the fact that prescriptions are issued mainly by general practitioners in primary care. This PhD thesis aimed at investigating the clinical effectiveness of independent pharmacist and diabetes specialist nurse (DSN) prescribers in the management of patients with type 2 diabetes at outpatient clinics in hospitals. A literature review was firstly conducted to explore the current research on NMP around the world and the UK. A systematic review of the previously published randomised control trials (RCT) and non-RCT studies that focused on prescribing interventions of nurses and pharmacist was also conducted to explore the impact of their prescribing interventions in treating type 2 diabetes using HbA1c level as the primary outcome. A programme of work of three retrospective comparative database analytical studies was then carried out to investigate the impact of independent NMPs in type 2 diabetes care. This programme of work used electronic medical records of patients attending outpatient clinics of diabetes centres in two teaching hospitals in Manchester; one employed an independent pharmacist and the other employed DSN prescribers. A group of subjects seen by an NMP in place of a doctor during the study period were the study group and the control group were those who seen only by doctors. The primary outcome was the average yearly change of HbA1c amongst the two groups. Secondary outcomes were yearly change of total cholesterol, blood pressure and serum creatinine as well as body mass index. Five statistical models, which included multivariable regression, propensity score matching and sensitivity analyses, were utilised to control for confounding effects, and the nature of selection bias in the retrospectively comparative effectiveness research using secondary database resources. A total of 330 patients seen by a team including a pharmacist versus 975 by doctors only between January 2006 and January 2013 at one site; and 656 by a team including DSNs versus 3,746 patients seen by doctors only between January 2007 to December 2013 at the other. The studies found both prescribing pharmacists and DSNs are capable of achieving at least non-inferior improvements in diabetes outcome compared to doctors. The pharmacist achieved a mean 0.01% reduction in HbA1c level versus doctors who achieved slight increase (p<0.4). DSNs also achieved a mean 0.07% reduction compared to doctors. However, after adjustment with multivariate and propensity score as well as with propensity score matching, there were no significant differences between the two groups. These findings were consistent with the findings in the systemic review. Although an RCT is the only method that by definition would produce unbiased treatment effects, the use of propensity score methods here, have reduced the potential for bias that may remain unaccounted for in multivariate models without propensity scores. Adjusting for propensity scores using two different methods also gives more confidence that the results are as unbiased as possible. Nonetheless, caution in generalising the results is necessary because of the retrospective nature of the studies and deficiencies in the database used.
156

Cooperativismo y su impacto en el rendimiento agropecuario local / Informe final PBA2AN51-233

Paredes Díaz, Carlos, Moreno Huaccha, Kenji 01 August 2019 (has links)
La presente investigación, ganadora del Concurso Anual de Investigación CIES 2018, cuenta con el auspicio de Global Affairs Canada (GAC), el Centro Internacional de Investigaciones para el Desarrollo (IDRC, por sus siglas en inglés), la Fundación Manuel J. Bustamante De la Fuente y la Superintendencia Nacional de Aduanas y de Administración Tributaria (SUNAT). / Este documento examina el impacto de la membresía cooperativa agropecuaria sobre los rendimientos financiero y productivo en los años 2015 y 2016. Empleamos los modelos Propensity Score Matching y Pooled OLS sobre los datos del Censo de Cooperativas y de la Encuesta Nacional Agropecuaria. Nuestro análisis revela que la membresía cooperativa afecta positivamente los rendimientos en los sectores agrario y agropecuario, pero no en el sector pecuario. Dicho impacto supera los impactos estimados para otras formas de asociatividad. Asimismo, se encuentra un efecto spillover positivo de una localidad con alta presencia cooperativa sobre localidades aledañas. Finalmente, se descubre que la participación femenina en directorios cooperativos agropecuarios también influye positivamente sobre los rendimientos de los productores agropecuarios. / This research examines the impact of agricultural cooperative membership on financial and productive performance in 2015 and 2016. We use the Propensity Score Matching and Pooled OLS models on the data from the Cooperative Census and the National Agricultural Survey. Our analysis reveals that cooperative membership positively affects yields in the agrarian and agricultural sectors, but not in the livestock sector. This impact exceeds the estimated impacts for other types of associativity. Likewise, there is a positive spillover effect of a locality with high cooperative presence over nearby localities yields. Finally, it is discovered that the participation of women in agricultural cooperative directories also positively influences the yields of agricultural producers.
157

An Examination of the Effects of Participation in High-Impact Practices Using Propensity Score Analysis with Structural Equation Modeling

Dickert, Joanna 08 December 2021 (has links)
No description available.
158

Finns det ett samband mellan födelseland och patientrapporterade utfall och upplevelser av vården efter stroke? : Jämförelse av statistiska metoder för att hantera skillnader i patientsammansättning / Is there a relationship between country of birth and patient reported outcome and experience of treatment after stroke? : Comparison of statistical methods for managing differences in case-mix

Ljungberg, Amanda, von Ahn, Agnieszka January 2021 (has links)
Studier visar att strokesjukvården i Sverige har en ojämn kvalitet. Denna studie syftar till att vidare undersöka möjliga ojämlikheter i strokevården. I uppsatsen undersöktes huruvida födelseland påverkar hur patienter som insjuknat i stroke skattar sin hälsa, nöjdhet med vården och nedstämdhet. Därtill jämfördes två statistiska metoder som kan hantera skillnader i patientsammansättning (t.ex. ålder och andra bakgrundsfaktorer), vilka var multipel logistisk regression och propensity score matching följt av konditionell logistisk regression. Båda metoderna gav signifikanta resultat. Resultatet visade att det är troligare att uppskatta sin allmänna hälsa som dålig, vara missnöjd med vården och vara nedstämd ofta eller ständigt för patienter födda utanför Sverige jämfört med de födda i Sverige, oberoende av vilken metod som användes. Propensity score matching visade större påverkan av födelseland på de undersökta responsvariablerna jämfört med multipel logistisk regression. Däremot var resultaten från multipel logistisk regression mer precisa.
159

COMPARISON OF SURVIVAL BETWEEN MELANOMA OF THE HEAD AND NECK AND MELANOMA OF OTHER REGIONS WITH PROPENSITY SCORE METHODS USING THE NATIONAL CANCER DATABASE

Cabrera Aviles, Claudia Ivette 23 May 2019 (has links)
No description available.
160

Effectiveness of the Pathways Community Hub Model in Reducing Low Birth Weight Among High-Risk Pregnant Women

Chiyaka, Edward Tafumaneyi 06 August 2019 (has links)
No description available.

Page generated in 0.257 seconds