<p>This dissertation consists of three chapters relating to</p> <p>identification and inference in dynamic microeconometric models</p> <p>including dynamic discrete games with many players, dynamic games with</p> <p>discrete and continuous choices, and semiparametric binary choice and</p> <p>duration panel data models.</p> <p>The first chapter provides a framework for estimating large-scale</p> <p>dynamic discrete choice models (both single- and multi-agent models)</p> <p>in continuous time. The advantage of working in continuous time is</p> <p>that state changes occur sequentially, rather than simultaneously,</p> <p>avoiding a substantial curse of dimensionality that arises in</p> <p>multi-agent settings. Eliminating this computational bottleneck is</p> <p>the key to providing a seamless link between estimating the model and</p> <p>performing post-estimation counterfactuals. While recently developed</p> <p>two-step estimation techniques have made it possible to estimate</p> <p>large-scale problems, solving for equilibria remains computationally</p> <p>challenging. In many cases, the models that applied researchers</p> <p>estimate do not match the models that are then used to perform</p> <p>counterfactuals. By modeling decisions in continuous time, we are able</p> <p>to take advantage of the recent advances in estimation while</p> <p>preserving a tight link between estimation and policy experiments. We</p> <p>also consider estimation in situations with imperfectly sampled data,</p> <p>such as when we do not observe the decision not to move, or when data</p> <p>is aggregated over time, such as when only discrete-time data are</p> <p>available at regularly spaced intervals. We illustrate the power of</p> <p>our framework using several large-scale Monte Carlo experiments.</p> <p>The second chapter considers semiparametric panel data binary choice</p> <p>and duration models with fixed effects. Such models are point</p> <p>identified when at least one regressor has full support on the real</p> <p>line. It is common in practice, however, to have only discrete or</p> <p>continuous, but possibly bounded, regressors. We focus on</p> <p>identification, estimation, and inference for the identified set in</p> <p>such cases, when the parameters of interest may only be partially</p> <p>identified. We develop a set of general results for</p> <p>criterion-function-based estimation and inference in partially</p> <p>identified models which can be applied to both regular and irregular</p> <p>models. We apply our general results first to a fixed effects binary</p> <p>choice panel data model where we obtain a sharp characterization of</p> <p>the identified set and propose a consistent set estimator,</p> <p>establishing its rate of convergence under different conditions.</p> <p>Rates arbitrarily close to <italic>n<super>-1/3</super></italic> are</p> <p>possible when a continuous, but possibly bounded, regressor is</p> <p>present. When all regressors are discrete the estimates converge</p> <p>arbitrarily fast to the identified set. We also propose a</p> <p>subsampling-based procedure for constructing confidence regions in the</p> <p>models we consider. Finally, we carry out a series of Monte Carlo</p> <p>experiments to illustrate and evaluate the proposed procedures. We</p> <p>also consider extensions to other fixed effects panel data models such</p> <p>as binary choice models with lagged dependent variables and duration</p> <p>models.</p> <p>The third chapter considers nonparametric identification of dynamic</p> <p>games of incomplete information in which players make both discrete</p> <p>and continuous choices. Such models are commonly used in applied work</p> <p>in industrial organization where, for example, firms make discrete</p> <p>entry and exit decisions followed by continuous investment decisions.</p> <p>We first review existing identification results for single agent</p> <p>dynamic discrete choice models before turning to single-agent models</p> <p>with an additional continuous choice variable and finally to</p> <p>multi-agent models with both discrete and continuous choices. We</p> <p>provide conditions for nonparametric identification of the utility</p> <p>function in both cases.</p> / Dissertation

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / This thesis brings two empirical studies using the methodologies of Sensitivity Analysis and Partial Identification on the causality of interventions of relevant importance to Brazilian education. These analyses are important tools, few times or never used in Brazilian literature, to investigate the questionable veracity of the assumptions used in the widespread Propensity Score Matching methodology. The first chapter expands the research on the performance gap between students from public and private schools with an analysis of Partial Identification. Proposed initially at Manski (1989), this methodology allows to question the assumptions commonly used in statistical and econometric models that points the private school system as more efficient. The second chapter investigates the effect of teacherâs stability on teaching quality. The goal is contributes to the Brazilian research on the characteristics of the teacher that affect student performance, inserting a nonmonetary factor related to hiring qualified professionals. Additionally, the estimative robustness will be verified by the Sensitivity Analysis proposed by Rosenbaum and Rubin (1983). The results indicate an overestimation of the positive effect of private school in studies of the Brazilian educational system and the significant impact of the stability condition of teachers on the quality of teaching in public schools. / Esta tese reÃne dois ensaios empÃricos utilizando as metodologias de AnÃlise de Sensibilidade e de IdentificaÃÃo Parcial sobre a relaÃÃo de causalidade de intervenÃÃes de relevante importÃncia na educaÃÃo brasileira. Essas anÃlises sÃo importantes ferramentas, pouco ou nunca utilizadas na literatura brasileira, para investigar a veracidade das questionÃveis suposiÃÃes utilizadas na difundida metodologia de Pareamento no Escore de PropensÃo. No primeiro capÃtulo à ampliada a pesquisa sobre a diferenÃa no desempenho dos alunos das escolas pÃblicas e privadas com a anÃlise de IdentificaÃÃo Parcial. Proposta inicialmente em Manski (1989), essa metodologia possibilita questionar as hipÃteses normalmente utilizadas nos modelos estatÃsticos e economÃtricos que conferem o sistema privado de ensino como o mais eficiente. No segundo capÃtulo investiga-se o efeito da estabilidade dos professores de escolas pÃblicas sobre a qualidade do ensino. Com o objetivo de contribuir com a pesquisa brasileira que investiga as caracterÃsticas do professor que afetam o desempenho dos alunos, inserindo um fator nÃo pecuniÃrio determinante na contrataÃÃo de profissionais qualificados. Adicionalmente, foi investigado a robustez das estimativas a partir da AnÃlise de Sensibilidade proposta por Rosenbaum e Rubin (1983). Os resultados revelam para a sobreestimaÃÃo do efeito positivo da escola privada nos estudos sobre o sistema de ensino brasileiro e o significativo impacto da condiÃÃo de estabilidade dos professores sobre a qualidade do ensino em escolas pÃblicas.

Cette thèse se compose de quatre travaux indépendants. Le premier concerne les modèles partiellement identifiés, c'est-à-dire des modèles dans lesquels la valeur du paramètre d’intérêt ne peut pas être déduite de la distribution des données et des hypothèses du modèle. Dans certaines situations, aucune ou au contraire plusieurs valeurs du paramètre d’intérêt sont compatibles avec les données et les hypothèses du modèle. Ce travail démontre entre autre que si l’ensemble des distributions de probabilités compatibles avec le modèle est convexe, alors les parties extrêmes de ce convexe caractérise l’ensemble des distributions compatibles avec le modèle. Le deuxième travail propose une méthode basée sur une condition d’exclusion pour corriger de l’attrition endogène dans les panels. Nous appliquons cette méthode pour estimer les transitions sur le marché du travail à partir de l’enquête emploi française. Le troisième travail propose une méthode simple pour estimer un modèle logistique avec effets fixes et dépendance d’état tel qu’étudié par Honoré et Kiriazidou. Il propose également un nouvel estimateur des écarts-types qui semble avoir de meilleures propriétés à distance finie. Le quatrième travail est une évaluation sur les collèges de la politique éducative des Réseaux-Ambition-Réussite lancée en 2006. Nous exploitons une discontinuité dans la sélection des collèges pour comparer entre eux certains collèges « identiques » avant la mise en place de la politique. Les résultats de cette évaluation laissent place à peu d’optimisme concernant l’efficacité de cette politique. / This work consists of four independent chapters. The first deals with partially identified models, i.e. models in which the parameter of interest is not uniquely identified from the data and the model’s assumptions. In some cases, no value of the parameter of interest is compatible with the data and the model’s assumption. And conversely in some cases, many values are compatible with the data and the model’s assumptions. This work demonstrates among other things that if the set of probability distributions compatible with the model is convex, then this set is characterized by its the extreme parts. The second paper proposes a method based on an exclusion restriction to correct endogenous attrition in panels. We apply this method to estimate the transitions on the labor market from the French Labour Force Survey. The third paper proposes a simple method to estimate a Logit model with fixed effects and state dependence as studied by Honoré and Kiriazidou. It also proposes a new method for inference. The fourth work is an evaluation of a French educational policy, implemented in 2006 whose objective is to provide extra resource to a subset of junior high schools. We exploit a discontinuity in treatment selection to build a credible counterfactual. Our results indicate some adverse treatment effects.

京都大学 / 0048 / 新制・課程博士 / 博士(経済学) / 甲第20870号 / 経博第565号 / 新制||経||283(附属図書館) / 京都大学大学院経済学研究科経済学専攻 / (主査)教授 西山 慶彦, 准教授 山田 憲, 准教授 高野 久紀 / 学位規則第4条第1項該当 / Doctor of Economics / Kyoto University / DGAM

This dissertation consists of three essays on Hedonic price method which is widely used in non-market good evaluation. The first chapter outlines three topics involved and briefly discusses the motivations and methods, as well we some conclusions in each of the following chapters. Chapter 2 uses a conventional first stage hedonic price method to estimate the effect of an aquatic invasive species (Eurasian watermilfoil) on lakefront property values at selected Vermont lakes. Results indicate that as the primary component of total aquatic macrophyte growth in a lake Eurasian watermilfoil significantly and substantially affects lakefront property values. As Eurasian watermilfoil infests a lake, adding to the total macrophyte growth, property values can diminish by <1% to 16% for incremental increases in the infestation level. Hence, policies that successfully prevent infestations have significant economic benefits to owners of lakefront properties and local communities. Chapter 3 focused on a previously unexplored potential impact of 9/11—the impact it may have had on housing prices near mosques. Using a unique dataset that combines the locations of functioning mosques with housing transactions near the time of 9/11, combined with a generalized difference-in-differences framework, we find that housing prices decreased by approximately 7% ($10,559 for the average home) in areas near mosques along the east coast of the U.S. on average in the two years following the attacks. However, on the west coast we find no evidence that 9/11 caused a systematic decrease in housing prices near mosques. Chapter 4 begins from a conventional model of hedonic equilibrium where a nonmarket amenity is conveyed as an attribute of a differentiated traded good. Different metropolitan areas may have different equilibrium price functions due to geographic variation in consumer preferences, income, and production costs. We demonstrate that under relatively mild restrictions on the geographic extent of taste-based sorting, indicator variables for metro areas define "imperfect instruments" that can be used to identify bounds on demand curves. Bounds on demand curves correspond to ranges of partial equilibrium welfare measures for non-marginal changes in environmental quality. We find these ranges to be informative in a preliminary application to evaluating the benefits of reducing cultural eutrophication of lakes in Maine, New Hampshire, and Vermont. The last chapter concludes and discusses the insights for future research. / Ph. D.

Dans ce mémoire, je considère un modèle de sélection standard avec sélection non aléatoire. D’abord, je discute la validité et la ‘‘sharpness ’’ des bornes sur l’intervalle interquantile de la distribution de la variable aléatoire latente non censurée, dérivées par Blundell et al. (2007). Ensuite, je dérive les bornes ‘‘sharp ’’ sur l’intervalle interquantile lorsque la distribution observée domine stochastiquement au premier ordre celle non observée. Enfin, je discute la ‘‘sharpness’’ des bornes sur la variance de la distribution de la variable latente, dérivées par Stoye (2010). Je montre que les bornes sont valides mais pas nécessairement ‘‘sharp’’. Je propose donc des bornes inférieures ‘‘sharp’’ pour la variance et le coefficient de variation de ladite distribution. / In this master thesis, I consider a standard selection model with nonrandomly censored outcome. First, I discuss validity and sharpness of bounds on the interquantile range of the distribution of the uncensored outcome, derived by Blundell et al. (2007). Second, I give sharp bounds on the interquantile range respectively under stochastic dominance of the unobserved outcome distribution by the observed one, and in presence of an exclusion variable. Third, I discuss sharpness of the variance bounds given by Stoye (2010). I show that the bounds are not necessarily sharp and I provide sharp lower bounds on the variance and the coe fficient of variation.

Submitted by Bernardo Andrade Lyrio Modenesi (bmodenesi@gmail.com) on 2015-07-22T21:31:40Z No. of bitstreams: 1 modenesi_thesis_v4whatseparatemen.pdf: 819403 bytes, checksum: d0a2f018fe44526173ae5b3f9e4c13a9 (MD5) / Rejected by Suzinei Teles Garcia Garcia (suzinei.garcia@fgv.br), reason: Prezado Bernardo bom dia, Nome da fundação e escola em negrito, o trabalho é em inglês vem primeiro o ABSTRACT depois o RESUMO e para aprovar o trabalho é necessário o seu orientador encaminhar e-mail para o cmcd@fgv.br com o novo Título autorizando o protocolo da versão final. Aguardo! Grata. Suzi 3799-7876 on 2015-07-23T11:30:51Z (GMT) / Submitted by Bernardo Andrade Lyrio Modenesi (bmodenesi@gmail.com) on 2015-07-23T14:56:25Z No. of bitstreams: 1 modenesi_thesis_v4.pdf: 819610 bytes, checksum: 8482a7c46b6c9bada5d33905e41147d7 (MD5) / Approved for entry into archive by Suzinei Teles Garcia Garcia (suzinei.garcia@fgv.br) on 2015-07-23T17:16:31Z (GMT) No. of bitstreams: 1 modenesi_thesis_v4.pdf: 819610 bytes, checksum: 8482a7c46b6c9bada5d33905e41147d7 (MD5) / Made available in DSpace on 2015-07-23T18:29:24Z (GMT). No. of bitstreams: 1 modenesi_thesis_v4.pdf: 819610 bytes, checksum: 8482a7c46b6c9bada5d33905e41147d7 (MD5) Previous issue date: 2015-06-25 / When estimating policy parameters, also known as treatment effects, the assignment to treatment mechanism almost always causes endogeneity and thus bias many of these policy parameters estimates. Additionally, heterogeneity in program impacts is more likely to be the norm than the exception for most social programs. In situations where these issues are present, the Marginal Treatment Effect (MTE) parameter estimation makes use of an instrument to avoid assignment bias and simultaneously to account for heterogeneous effects throughout individuals. Although this parameter is point identified in the literature, the assumptions required for identification may be strong. Given that, we use weaker assumptions in order to partially identify the MTE, i.e. to stablish a methodology for MTE bounds estimation, implementing it computationally and showing results from Monte Carlo simulations. The partial identification we perfom requires the MTE to be a monotone function over the propensity score, which is a reasonable assumption on several economics' examples, and the simulation results shows it is possible to get informative even in restricted cases where point identification is lost. Additionally, in situations where estimated bounds are not informative and the traditional point identification is lost, we suggest a more generic method to point estimate MTE using the Moore-Penrose Pseudo-Invese Matrix, achieving better results than traditional methods. / A estimação de parâmetros relevantes no contexto de intervenções políticas, também conhecidos como efeitos de tratamento, enfrenta diversos problemas práticos como o viés relacionado ao mecanismo de atribuição do status de tratamento. Adicionalmente, efeitos heterogênos na literatura de Avaliação de Impactos é mais comum que efeitos homogêneos. Em situações nas quais estes problemas estão presentes, a estimação do Efeito Marginal de Tratamento (MTE) emprega o método de variáveis instrumentais para contornar o viés de seleção ao tratamento, obtendo ao mesmo tempo efeitos heterogêneos. Mesmo que a literatura identifique pontualmente este parâmetro, as hipóteses por trás da identificação são costumeiramente restritivas. Neste trabalho pretende-se afrouxar as hipóteses vigentes na literatura de modo a obter identificação parcial do MTE, requerendo apenas monotonicidade do mesmo ao longo das diferentes propensões ao tratamento, o que é comum em diversos exemplos da literatura econômica. Simulações de Monte Carlo são performadas, resultando em limites para o MTE que se mostram informativos, mesmo em situações restritas nas quais a tradicional identificação pontual é perdida. Complementarmente, em situações quando a identificação parcial não é informativa e a tradicional ponto identificação é perdida, propomos a ponto identificação utilizando a matrix pseudo inversa de Moore-Penrose. Esta metodologia prescinde da hipótese de monotonicidade e resulta em melhores estimativas quando comparada aos métodos tradicionais.