Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-07-26T20:15:21Z No. of bitstreams: 1 esthergrizendegarcia.pdf: 2061658 bytes, checksum: 362dbb4c657f5be1883e9421ad02ba40 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-07-27T11:32:10Z (GMT) No. of bitstreams: 1 esthergrizendegarcia.pdf: 2061658 bytes, checksum: 362dbb4c657f5be1883e9421ad02ba40 (MD5) / Made available in DSpace on 2016-07-27T11:32:10Z (GMT). No. of bitstreams: 1 esthergrizendegarcia.pdf: 2061658 bytes, checksum: 362dbb4c657f5be1883e9421ad02ba40 (MD5) Previous issue date: 2016-06-20 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / FAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Gerais / O presente trabalho tem como objetivo investigar a relação entre o estado de saúde dos indivíduos e seus rendimentos provenientes do trabalho. Indivíduos doentes podem perder produtividade em sua atividade profissional, podem necessitar se ausentar do trabalho por um período, ou até mesmo deixar de trabalhar. O estudo é realizado para o Brasil, utilizando os dados da PNAD de 2003 e 2008, que contêm o suplemento de saúde. Para a pesquisa foram selecionados apenas os indivíduos do gênero masculino com idade entre 18 e 65 anos. São criadas as variáveis estado de saúde e nível socioeconômico, a partir da Análise do Componente Principal (PCA). Para a investigação, utiliza-se o método Propensity Score Matching (Escore de Propensão). Os indivíduos não saudáveis (grupo de tratamento) estão pareados com um subgrupo de indivíduos saudáveis (grupo de controle) a fim de obter atributos semelhantes de ambos os grupos, em que a única característica que os diferenciam é a condição de saúde. O estudo utiliza três tipos de pareamento: Vizinho Mais Próximo (Nearest Neighbor Matching), Pareamento por Raio (Radius Matching) e Pareamento por Kernel. Em seguida, utiliza-se o modelo minceriano para mensurar o retorno salarial. Os resultados da pesquisa demostram que o estado de saúde é relevante para explicar rendimentos do trabalho principal. Considerando os três métodos de pareamento, os indivíduos não saudáveis auferem em torno de 13,2% a 21% a menos que os indivíduos saudáveis. Esse efeito é corroborado quando os resultados são estimados pela combinação do método de pareamento com regressão linear, indivíduos não saudáveis auferem, em média, rendimentos entre 11,75% e 15,9% menores que os rendimentos de indivíduos saudáveis. / This study aims to investigate the relationship between the health status of individuals and their income from work. Ill patients may lose productivity in their professional activity, they may need to be absent from work for a period, or even stop working. The study is conducted for Brazil, using the PNAD data of 2003 and 2008, which contain the health supplement. For the research were selected only male subjects aged between 18 and 65 years. The variables health and socioeconomic status, from the Principal Component Analysis (PCA) are created. For research, we use the propensity score matching method (Score Propensity). Non-healthy individuals (treatment group) are matched with a subset of healthy individuals (control group) to obtain similar attributes of both groups, wherein the unique feature that distinguish them is the health condition. The study uses three types of pairing: Nearest Neighbor (Nearest Neighbor Matching), pairing by Radius (Radius Matching) and pairing by Kernel. Then you use the mincerian model to measure the wage returns. The survey results demonstrate that the health status is relevant to explain income of the main job. The survey results show that, considering the three methods of matching, unhealthy individuals earn 13.2% to 21% less than healthy individuals. This effect is higher considering the regression model combined with matching, unhealthy individuals earn, on average, around 11.75% and 15.9% lower than the healthy individuals.

Background: Little is known regarding the health-related quality of life among myocardial infarction (MI) survivors in the United States. The purpose of this population-based study was to identify differences in health-related quality of life domains between MI survivors and propensity score matched controls. Methods: This retrospective, cross-sectional matched case-control study examined differences in health-related quality of life (HRQoL) among MI survivors of myocardial infarction compared to propensity score matched controls using data from the 2015 Behavioral Risk Factor Surveillance System (BRFSS) survey. Propensity scores were generated via logistic regression for MI survivors and controls based on gender, race/ethnicity, age, body mass index (BMI), smoking status, and comorbidities. Chi-square tests were used to compare differences between MI survivors to controls for demographic variables. A multivariate analysis of HRQoL domains estimated odds ratios. Life satisfaction, sleep quality, and activity limitations were estimated using binary logistic regression. Social support, perceived general health, perceived physical health, and perceived mental health were estimated using multinomial logistic regression. Significance was set at p < 0.05. Results: The final sample consisted of 16,729 MI survivors matched to 50,187 controls (n = 66,916). Survivors were approximately 2.7 times more likely to report fair/poor general health compared to control (AOR = 2.72, 95% CI: 2. 43-3.05) and 1.5 times more likely to report limitations to daily activities (AOR = 1.46, 95% CI: 1.34-1.59). Survivors were more likely to report poor physical health > 15 days in the month (AOR = 1.63, 95% CI: 1.46-1.83) and poor mental health > 15 days in the month (AOR = 1.25, 95% CI: 1.07-1.46) compared to matched controls. There was no difference in survivors compared to controls in level of emotional support (rarely/never: AOR = 0.75, 95% CI: 0.48-1. 18; sometimes: AOR = 0.73, 95% CI: 0.41-1.28), hours of recommended sleep (AOR = 1.14, 95% CI: 0.94-1.38), or life satisfaction (AOR = 1.62, 95% CI: 0.99-2.63). Conclusion: MI survivors experienced lower HRQoL on domains of general health, physical health, daily activity, and mental health compared to the general population.

The Zambian government, through the Ministry of Agriculture and Cooperatives (MACO), provides maize seed and fertilizers to farmers at heavily subsidised prices under the Farmer Input Support Programme (FISP). MACO’s narrow evaluation of FISP, based on estimated production without quantifying the significant changes in production and other critical socioeconomic factors, fails to adequately highlight and service the benefits of subsidies to intended beneficiaries. Furthermore, MACO estimates of the impact of FISP never consider the question of how much beneficiary farmers would have produced in its absence, leading to potentially misleading assessments. The key question addressed in this study is whether using more rigorous econometric methods that account for heterogeneity in socioeconomic factors between participants and non-participants would still confirm the positive impact of FISP on maize productivity and poverty reduction, hence justifying the huge government expense on the programme. The study utilised cross-sectional data obtained from 497 randomly selected households, collected in 2011 from six provinces of Zambia to assess the causal effect of FISP on beneficiary households’ maize yields and asset accumulation. The data was analysed using well-grounded matching techniques that account for differences in observable characteristics between programme participants and non-participants. The study also tested for possible unobserved selection effects using the Rosenbaum bounds. The results indicated that participating in FISP increased maize yields and assets accumulation and hence might directly or indirectly positively affect beneficiary poverty levels. There were also no influences of unobserved characteristics on the estimated maize yield and asset level differences between participants and non-participants. On average, FISP increased maize yields by about 451 kg per hectare, with an improvement of about 0.5 on the wealth index (score used to rank households according to asset levels). The positive impact on maize yields and asset accumulation on the participating farmers therefore justifies government’s continued implementation of FISP. / Dissertation (MScAgric)--University of Pretoria, 2014. / lk2014 / Agricultural Economics, Extension and Rural Development / MScAgric / Unrestricted

I examine the impact of a health insurance program for the poor implemented in Indonesia in 2005 on labor supply and informal work measured by employment status outside of the private and public formal sector. As a first step to its ambitious plan for universal coverage, this program extended subsidized health insurance coverage to a large proportion of Indonesia’s poor and near-poor population. Using a rich longitudinal survey, I estimate the average treatment effect on the treated using a combined propensity score matching method with difference in differences. The richness of the data allows me to control for a set of observable characteristics used by the government to allocate the benefit as well as an extensive combination of controls at the individual, household and community level. I find a significant negative impact on labor supply. This impact is driven by women, both at the intensive and extensive margin of labor supply. These results are in line with the fact that individuals with higher value for health insurance are more inclined to modify their labor market behavior. This decrease in labor supply has important policy implications as it may cause a negative impact on economic development, poverty and socioeconomic status of women. The results are not suggestive of an impact on informality. This lack of an effect on informal status is encouraging for developing countries in the verge of implementing universal care reform.

This note proposes the continuous treatment approach as a valuable alternative to propensity score matching for evaluating economic effects of merger and acquisitions (M&As). This framework allows considering the variation in treatment intensities explicitly, and it does not call for an arbitrary definition of cutoff values in traded ownership shares to construct a binary treatment indicator. We demonstrate the usefulness of this approach using data from European M&As and by relying on the example of post-M&A employment effects. The empirical exercise reveals some heterogeneities over the whole distribution of acquired ownership shares and across different types of M&As and country groups.

By means of data-driven marketing as well as big data technology, this paper presents the investigation of a case study from travel industry implemented by a combination of propensity model and a business model “2W1H”. The business model “2W1H” represents the purchasing behavior “What to buy”, “When to buy”, and “How to buy”. This paper presents the process of building propensity models for the application in behavioral targeting in travel industry.     Combined the propensity scores from predictive analysis and logistic regression with proper marketing and CRM strategies when communicating with travelers, the business model “2W1H” can perform personalized targeting for evaluating of marketing strategy and performance. By analyzing the business model “2W1H” and the propensity model on each business model, both the validation of the model based on training model and test data set, and the validation of actual marketing activities, it has been proven that predictive analytics plays a vital role in the implementation of travelers’ purchasing behavioral targeting in marketing.

Magíster en Gestión de Operaciones. Ingeniero Civil Industrial / Con el crecimiento del uso de smartphones e internet móvil, se ha producido un aumento en el uso de aplicaciones, y ante ello la publicidad móvil aparece como uno de los mercados más importantes en la industria. Una de las posibilidades que entrega este medio de comunicación, es el aprovechamiento de la geo localización como herramienta de targeting, permitiendo entregar publicidad en función del contexto en que se encuentra el usuario, además de identificarlo a partir de la información ingresada y su interacción con la aplicación. El presente trabajo consiste en realizar un modelo de decisión para una aplicación que envía promociones geo localizadas de distintas marcas a sus usuarios, con el objetivo de asignar promociones en función de su probabilidad de apertura y rentabilidad. Para ello se generó un set de datos, donde las variables predictivas fueron clasificadas en grupos según la información que contenían, esto es, información sobre el usuario, la promoción, la marca que envía la promoción, la relación del usuario con la aplicación y con la marca, y el tiempo. A partir de este set de datos se busca generar un modelo de propensión y en base a él, crear un modelo que tome decisiones respecto al envío de promociones. Para calcular la probabilidad de apertura de cada envío, se utilizó un enfoque econométrico, utilizando una regresión logística como algoritmo predictivo. A través de los análisis y sus resultados, se descubrieron diferencias significativas en el poder predictivo de los distintos grupos de variables, según la edad de los usuarios y la cantidad de promociones que han recibido. Por lo tanto, para internalizar estas diferencias en el modelo, se probaron modelos heterogéneos, así como interacciones entre variables. Para el modelo de decisión en tanto, se utilizó un enfoque económico que permita beneficiar a la compañía y a sus usuarios. A partir de los resultados obtenidos se concluye que la interacción entre variables permite tener mejoras en el poder predictivo del modelo, sin complejizar la implementación del sistema a diferencia de los modelos heterogéneos. De esta manera, la preferencia de marcas varía según la edad de los usuarios y la importancia de la geo localización es más relevante en las primeras etapas del usuario. Finalmente, a través de una simulación, se comprobó que el modelo permite tener mejoras en la asignación de promociones, incrementando la tasa de apertura, y permitiendo aumentar las utilidades de la compañía, al disminuir el envío de promociones no relevantes y no rentables. Además, el modelo permite comprender los factores que inciden en la apertura, estimar el resultado de las campañas y automatizar el proceso de segmentación para el envío de promociones. / 28/11/2022

Background: Current cesarean delivery (CD) rates in many industrialized countries are well above the recommended rates. Objective: The overall goal of this thesis was to identify sources for unnecessary CD. Specific objectives were to: 1) analyze the leading indications for CD and their associations with neonatal outcomes; 2) compare adverse birth outcomes between elective primary cesarean delivery (EPCD) and trial of labor after vaginal birth (TOLAV), and between elective repeat cesarean delivery (ERCD) and trial of labor after cesarean birth (TOLAC); and 3) assess the cost-effectiveness of ERCD and TOLAC. Methods: A retrospective cohort study was conducted. Leading indications for CD were analyzed and risks of neonatal outcomes between “soft” indications and “hard” indications were compared first, using 2006 to 2013 Better Outcomes and Registry Network Ontario data. A pair of analyses: comparing risks of adverse birth outcomes between EPCD and TOLAV and between ERCD and TOLAC, were then conducted using United States 2005 to 2010 birth registration data. Analysis were performed using logistic regression and propensity score matching models. Finally, a cost-effectiveness analysis between ERCD and TOLAC was performed. Results: The single largest contributor for overall CD was ERCD (34.3%) and for primary CD was dystocia (31.9%) in Ontario. Compared with infants of mothers with CD for “hard” indications, the risks of Apgar score <7 at 5 minutes for infants of mothers with CD for non-reassuring-fetal-status was increased, while the risks of Apgar score <7 at 5 minutes and neonatal death for infants of mothers with ERCD and dystocia were decreased. Compared with infants of mothers who underwent TOLAV, infants of mothers who underwent EPCD were more likely to require antibiotics and ventilation support, but less likely to have birth injury. On the other hand, compared with infants of mothers who underwent TOLAC, infants of mothers who underwent ERCD were less likely to require antibiotics and ventilation support. ERCD was similar to the TOLAC birth option in terms of cost effectiveness. Conclusions: Tight up criteria for “soft” indications such as labor dystocia could result in substantial reduction in CD without harming the infants.

Non-inferiority tests are well developed for randomized parallel group trials where the control and experimental groups are independent. However, these tests may not be appropriate for assessing non-inferiority in correlated one-to-many matched data. We propose a new statistical test that extends Farrington-Manning’s (FM) test to the case where many (≥1) control subjects are matched to each experimental subject. We conducted a Monte Carlo simulation study to compare the size and power of the proposed test with tests developed for clustered one-to-one matched pair data and tests based on generalized estimating equations (GEE). For various correlation patterns, the sizes of tests developed for clustered matched pair data and GEE-based tests are inflated when applied to the case where many control subjects are matched to each experimental subject. The size of the proposed test, on the other hand, is close to the nominal level for a variety of correlation patterns. There is a debate in the literature regarding whether or not statistical tests appropriate for independent samples can be used to assess the statistical significance of treatment effects in propensity-score matched studies. We used Monte Carlo simulations to examine the effect on assessing non-inferiority via risk difference when a method for independent samples (i.e. FM test) is used versus when a method for correlated matched samples is used in propensity-score one-to-many matched studies. If propensity-score matched samples are well-matched on baseline covariates and contain almost all of the experimental treated subjects, a method for correlated matched samples is preferable with respect to power and Type I error than a method for independent samples. Sometimes there are more experimental subjects to choose from for matching than control subjects. We conducted a Monte Carlo simulation study to compare the size and power of the previously mentioned tests when many (≥1) experimental subjects are matched to each control subject. In this case, the Nam-Kwon test for clustered data performs the best in controlling the type I error rate for a variety of correlation patterns. Therefore, the appropriate non-inferiority test to use for correlated matched data depends, in part, on the sample size allocation of subjects.

Thesis advisor: Henry I. Braun / Sponsored by OECD, PIAAC represents the first attempt to assess adult problem solving in technology-rich environments (PS-TRE) on an international scale that is comparable cross-culturally and cross-nationally. The objectives of this study are to study (1) the distributions of PS-TRE proficiency scores across 14 selected countries and (2) within each country, the associations between PS-TRE proficiency scores and the different formats of adult education and training (AET) participation. Using data on full-time professionals (at least 25 years old) from these countries, propensity score weighting was applied to estimate the associations between the different formats of AET participation and their PS-TRE proficiency scores. To place these estimates in context, parallel analyses were conducted – one with the sample of full-time associates in the 14 selected countries and the other with full-time professionals’ Literacy and Numeracy proficiency scores as measured by PIAAC. The results showed that after controlling for socio-demographic background, occupational categories, use of key information-processing skills (both at home and at work), as well as use of generic workplace skills, no consistent pattern was found across the 14 selected countries. At the individual country level, scattered significant relationships were identified. For example, in Denmark, both formats of AET participation (vs. None) are significantly and positively associated with full-time professionals’ PS-TRE proficiency scores and their probability of scoring in the top quartile of the PS-TRE distribution (p < .01). While in the United States, Formal AET (vs. None) is significantly and positively associated with full-time associates’ PS-TRE proficiency scores and their probability of scoring in the top quartile of the PS-TRE distribution (p < .01). The variations in relationships between the different formats of AET participation and working adults’ skills proficiency across domains and samples indicate the necessity of conducting qualitative research on AET programs in individual countries. Furthermore, to provide recommendations tailored to the specific needs of each country, a fine-grained classification of AET programs based on the OECD guideline was suggested. / Thesis (PhD) — Boston College, 2020. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Research, Measurement and Evaluation.