A common problem in regression analysis is that some covariates are measured with errors. In this dissertation we present simulation-based approach to estimation in two popular regression models with a categorical response variable and classical measurement errors in covariates. The first model is the regression model with a binary response variable. The second one is the proportional odds regression with an ordinal response variable. In both regression models we consider method of moments estimators for therein unknown parameters that are defined via minimizing respective objective functions. The later functions involve multiple integrals and make obtaining of such estimators unfeasible. To overcome this computational difficulty, we propose Simulation-Based Estimators (SBE). This method does not require parametric assumptions for the distributions of the unobserved covariates and error components. We prove consistency and asymptotic normality of the proposed SBE's under some regularity conditions. We also examine the performance of the SBE's in finite-sample situations through simulation studies and two real data sets: the data set from the AIDS Clinical Trial Group (ACTG175) study for our logistic and probit regression models and one from the Adult Literacy and Life Skills (ALL) Survey for our regression model with the ordinal response variable and mismeasured covariates. / October 2016

The upward trend in the prevalence of childhood mental disorders observed in the UK over the previous two decades, together with UK’s poor performance in recent international comparisons of child well-being, has brought childhood social and emotional well-being (SEW) to the forefront of policy. Key to tackling this issue is to understand what causes SEW in childhood, what interventions are successful in improving it, and what are its late-life consequences. This thesis furthers the literature in each of these areas. Firstly, we examine whether foetal (or in-utero) exposure to influenza hampers the development of childhood SEW. To do so, we examine the use of an instrumental variables approach, whereby the severity of the 1957 Asian Flu epidemic in the local authority of birth is used as an instrument for whether mothers self-report contracting influenza during pregnancy. We establish that exposure has little effect on childhood SEW, but that it results in a 60% increase in the risk of being stillborn, suggesting an increasing focus on influenza vaccination during pregnancy is needed. Secondly, we investigate the long-term effectiveness of school-based interventions to improve SEW. In order to overcome the absence of long follow-up in trial datasets, we develop a new modelling approach which involves the matching of trial participants to individuals in birth cohort datasets. An application of this method found that a Promoting Alternative Thinking Strategies (PATHS) intervention implemented in Greater Manchester schools led to a statistically significant improvement in childhood SEW, and had a positive, although statistically insignificant, effect on health across the life-course. Finally, we address the paucity of studies examining the effects of childhood SEW on late-life health and labour market outcomes. To do so, we develop a method for generating predictions of the effects of childhood characteristics beyond the currently available follow-up periods in birth cohort datasets, adapting an existing mediation analysis framework. Applying this method, we establish that a one-standard deviation improvement in childhood SEW leads to an increase of up to 0.18 accumulated quality-adjusted-life-years in late-life, and an increase in pre-tax labour income in late-life of up to £23,850. Both of these effects are primarily driven by large positive effects of childhood SEW on educational attainment, employment, income and health in mid-life. Childhood SEW is a predictor of important outcomes throughout the lifecourse. More research is needed to identify its causes and interventions to successfully improve it.

Submitted by Maurício Chikitani (mauricio.chikitani@gmail.com) on 2015-05-05T12:12:27Z No. of bitstreams: 1 tese_abnt.pdf: 585315 bytes, checksum: 483f7c0bad34b15345e4103192a166aa (MD5) / Approved for entry into archive by Suzinei Teles Garcia Garcia (suzinei.garcia@fgv.br) on 2015-05-05T12:20:03Z (GMT) No. of bitstreams: 1 tese_abnt.pdf: 585315 bytes, checksum: 483f7c0bad34b15345e4103192a166aa (MD5) / Made available in DSpace on 2015-05-05T13:02:25Z (GMT). No. of bitstreams: 1 tese_abnt.pdf: 585315 bytes, checksum: 483f7c0bad34b15345e4103192a166aa (MD5) Previous issue date: 2015-04-29 / Eu analiso o papel que a interação entre pares tem na determinação do Locus de Controle, uma medida de quão pessoalmente responsável as pessoas se sentem com relação a diferentes aspectos de suas vidas. Eu estabeleço identificação através de variação no tamanho do grupo e de variáveis instrumentais baseadas na estrutura de painel dos dados. Eu estudo a questão no contexto escolar, utilizando os dados de alunos no Ensino Fundamental de um município brasileiro, que inclui o questionário de Tel Aviv de Locus de Controle. Minhas estimativas não apresentam sinais de peer effects endógenos ou contextuais. / I analyze the role of peer interaction in the determination of the Locus of Control, a measure of how personally responsible people feel about their life affairs. Identification is established using group size variation and instrumental variables based on the panel structure of the data. I study this question on an educational setup, using Middle School data in a Brazilian municipality that includes the Tel Aviv Locus of Control questionnaire. My estimates show no sign of endogenous nor contextual peer effects.

In recent years the program evaluation methods have become very popular in applied microeconomics. However, the variety of these methods responds to specific problems, which are normally determined by the data available and the impact the researcher tries to measure. This paper summarizes the main methods in the current literature, emphasizing the assumptions under which the average treatment effect and the average treatment effect on the treated are identified. Additionally, after each section I briefly present some applications of these methods. This document is a didactic presentation for advanced students in economics and applied researchers who wish to learn the basics of these techniques / En años recientes los métodos de evaluación de impacto se han difundido ampliamente en la investigaciónmicroeconómica aplicada. Sin embargo, la variedad de métodos responde a problemas particulares y específicos los cuales están determinados normalmente por los datos disponibles y el impacto que se busca medir. El presente documento resume las principales corrientes disponibles en la literatura actual, poniendo énfasis en los supuestos bajo los cuales el efecto tratamiento promedio y el efecto tratamiento promedio sobre los tratados se encuentran identificados. Adicionalmente se presentan algunos ejemplos de aplicaciones prácticas de estos métodos. Se busca hacer una presentación didáctica que pueda ser útil a estudiantes avanzados y a investigadores aplicados que busquen conocer los principios básicos de estas técnicas.

nÃo hà / This work aims to analyze the impact of education on wage of individuals in the state of CearÃ. To this end, it was estimated mincerian equations, seeking the school returns for years 2003, 2006 and 2009. The data used was the PNAD, developmente by IBGE. Two Instrumental variables, that aim correct the problem of endogeneity in this kind of estimation: 1) the first, it was used parental education as instrument, aiming at correcting the problem of family background; 2) the second, used was instruments the wife education and experience, based on methodology development by Di Pietro and Pedace (2007). The results agree with recent literature, showing that Cearaâs state the school returns is less than national average. / Este trabalho tem o objetivo de analisar o impacto da educaÃÃo no salÃrio dos indivÃduos no estado do CearÃ. Para tanto, estimou-se equaÃÃes mincerianas, buscando-se os retornos escolares para os anos de 2003, 2006 e 2009. A base de dados utilizada foi a Pesquisa Nacional de Amostragem por Domicilios (PNAD), desenvolvida pelo IBGE. Dois tipos de variÃveis instrumentais, que visam corrigir o problema da endogenia presente nesse tipo de estimaÃÃo, foram utilizadas: 1) A primeira, cujo instrumento à a escolaridade dos pais, visa corrigir o problema da heranÃa familiar; 2) A segunda, cujo instrumento à a escolaridade e a experiÃncia das esposas, com base em metodologia desenvolvida por Di Pietro e Pedace (2007). Os resultados corroboraram a literatura recente, indicando que o Estado do Cearà possui retornos educacionais mÃdios abaixo da mÃdia nacional.

During fast point-to-point motion, the inherent joint flexibility could be detrimental in terms of residual vibration. Aiming to minimize the vibration, the command shaping method has been developed so as to remove critical energy from the input profile at resonant frequencies. Since this method requires information of a physical model in order to find the target frequencies, the quality of the shaped command profile relies on the accuracy of the model parameter estimation. Therefore, in this work, a system identification method using Instrumental Variables is applied from the literature. Compared with the classic Ordinary Least Square method, the IV approach has successfully improved the estimation of parameters, based on simulation results. The accuracy of parameter estimation influences the command profile, as does the feedback controller. In this work, starting from a mathematical derivation with a mismatch model due to a feedback controller called Computed Torque Control, insight for the closed-loop system is given with regard to the interaction between control gains and the actual resonant frequencies. It is found that the control gain is able to modify the actual resonant frequency curve, and push it into or out of the shaping bounds which are generated from the command shaping method. Further analysis based on the simulation results shows that the overlap area between the shaping bounds and the actual frequencies affects the level of residual vibration. In light of this fact, an optimal control gain exists and is found when the estimation error is in a certain range. At the end, recommendations for choosing the control gains are provided.

No description available.

Indiana University-Purdue University Indianapolis (IUPUI) / Randomized studies are designed to estimate the average treatment effect (ATE) of an intervention. Individuals may derive quantitatively, or even qualitatively, different effects from the ATE, which is called the heterogeneity of treatment effect. It is important to detect the existence of heterogeneity in the treatment responses, and identify the different sub-populations. Two corresponding statistical methods will be discussed in this talk: a hypothesis testing procedure and a mixture-model based approach. The hypothesis testing procedure was constructed to test for the existence of a treatment effect in sub-populations. The test is nonparametric, and can be applied to all types of outcome measures. A key innovation of this test is to build stochastic search into the test statistic to detect signals that may not be linearly related to the multiple covariates. Simulations were performed to compare the proposed test with existing methods. Power calculation strategy was also developed for the proposed test at the design stage. The mixture-model based approach was developed to identify and study the sub-populations with different treatment effects from an intervention. A latent binary variable was used to indicate whether or not a subject was in a sub-population with average treatment benefit. The mixture-model combines a logistic formulation of the latent variable with proportional hazards models. The parameters in the mixture-model were estimated by the EM algorithm. The properties of the estimators were then studied by the simulations. Finally, all above methods were applied to a real randomized study in a low ejection fraction population that compared the Implantable Cardioverter Defibrillator (ICD) with conventional medical therapy in reducing total mortality.

This thesis investigates the relationship between smallholder farm household networks for food acquisition and agricultural production, food security and dietary quality in the Mount Elgon region of western Kenya and eastern Uganda. Food security and dietary quality were measured through calorie consumption of the female household head in a 24 hour dietary recall, the calculation the World Food Program Food Consumption Score (WFP FCS), and the calculation of the percentage of energy sourced from staples in the diet. Correlations between these indicators support that the WFP FCS is capturing elements of both sufficiency and quality of diet. Subsequent application of Ordinary Least Squares regression determines that both food acquisition networks and technology networks for agricultural production have a statistically significant positive impact upon calorie procurement across the sites included in the study. However, networks for agricultural production appear to operate differently in different locations with regard to dietary quality. Interpretation of qualitative data gathered through interviews with agricultural service sector providers and focus groups regarding these local networks for agricultural production suggests that this might be due to differences in the types of crops promoted and attitudes held regarding food security and dietary quality prevalent in these different localities. Overall, the results suggest that both food acquisition networks and agricultural production networks are important avenues through which gains in food security may be realized. However, development efforts need to be mindful of the crops and attitudes promoted by these networks to secure gains in both caloric sufficiency and dietary quality. / Master of Science / CCRA-8 (Technology Networks for Sustainable Innovation)

Effects of human capital in developing countries Bc. Jan Mouček Abstract Improving education and health of people living in developing countries is an often discussed topic. We however know relatively little about the effects that investments into schooling and improving the nutrition and hygiene of people in those countries have on their productivity at work. This thesis investigates the effects of four variables representing investment into human capital on income of individuals from Bangladesh and Indonesia. The four studied variables are education, migration, height and Body Mass Index. The thesis aims at comparing results of estimation using instrumental variables estimator with previous similar research. It tries to observe possible differences in size of effects the human capital variables have. It finds however that the instrumental variables used are of insufficient quality for the estimation to yield reliable results. Because the instruments used were almost the same as in underlying works and their quality has not been previously tested the conclusion of this thesis is that further search for correct instrumental variables describing the human capital variables in various countries around the globe is necessary.