Indiana University-Purdue University Indianapolis (IUPUI) / Observational data are frequently used for causal inference of treatment effects on prespecified outcomes. Several widely used causal inference methods have adopted the method of inverse propensity score weighting (IPW) to alleviate the in uence of confounding. However, the IPW-type methods, including the doubly robust methods, are prone to large variation in the estimation of causal e ects due to possible extreme weights. In this research, we developed an ordinary least-squares (OLS)-based causal inference method, which does not involve the inverse weighting of the individual propensity scores. We first considered the scenario of homogeneous treatment effect. We proposed a two-stage estimation procedure, which leads to a model-free estimator of average treatment effect (ATE). At the first stage, two summary scores, the propensity and mean scores, are estimated nonparametrically using regression splines. The targeted ATE is obtained as a plug-in estimator that has a closed form expression. Our simulation studies showed that this model-free estimator of ATE is consistent, asymptotically normal and has superior operational characteristics in comparison to the widely used IPW-type methods. We then extended our method to the scenario of heterogeneous treatment effects, by adding in an additional stage of modeling the covariate-specific treatment effect function nonparametrically while maintaining the model-free feature, and the simplicity of OLS-based estimation. The estimated covariate-specific function serves as an intermediate step in the estimation of ATE and thus can be utilized to study the treatment effect heterogeneity. We discussed ways of using advanced machine learning techniques in the proposed method to accommodate high dimensional covariates. We applied the proposed method to a case study evaluating the effect of early combination of biologic & non-biologic disease-modifying antirheumatic drugs (DMARDs) compared to step-up treatment plan in children with newly onset of juvenile idiopathic arthritis disease (JIA). The proposed method gives strong evidence of significant effect of early combination at 0:05 level. On average early aggressive use of biologic DMARDs leads to around 1:2 to 1:7 more reduction in clinical juvenile disease activity score at 6-month than the step-up plan for treating JIA.

Orientador: José Plínio de Oliveira Santos / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matemática Estatística e Computação Científica / Made available in DSpace on 2018-08-27T11:40:03Z (GMT). No. of bitstreams: 1 Nascimento_MarcusViniciusSilva_M.pdf: 918557 bytes, checksum: de0f1627892732c764e7f5046966336f (MD5) Previous issue date: 2015 / Resumo: Nosso objetivo, nesse trabalho, é o de fazer um estudo sobre o método do crivo. A motivação reside no desejo de aplicar essas ideias a uma situação particular. Dividimos nosso trabalho em três partes. Na primeira fornecemos apenas as definições e con- ceitos básicos. Na segunda apresentamos o principio da inclusão-exclusão que embora sendo algo bastante conhecido merece destaque especial dada a sua importância como ferramenta no nosso trabalho. Na terceira e última parte, fazemos uma contextualização histórica e uma descrição da evolução das ideias do crivo de Eratóstenes-Legendre. A escolha desse crivo, dentre tantos outros, foi feita tendo em vista dois pontos. O primeiro é que o crivo de Eratóstenes-Legendre é o mais simples dentre os crivos estudados na teoria dos crivos. O segundo ponto está relacionado com o fato deste crivo fornecer a ideia geral dos crivos combinatoriais, uma vez que os crivos mais sofisticados são extensões de suas ideias básicas / Abstract: Our aim in this work is to make a study about the sieve method. The motivation lies in the intent of applying this idea in a particular situation. We splitted the study into three parts. The first part deals with definitions and basic concepts. In the second we present the principle of inclusion-exclusion while being something well known deserves special mention given its importance as a tool in our work. In the third and final part, we make a historical contextualization and a description of the evolution of the sieve Eratosthenes- Legendre ideas. The choice of sieve, among many others, has been made taking into account two points. The first is that the Eratosthenes-Legendre sieve is the simplest among the sieves studied the theory of sieves. The second point is related to the fact that this sieve provide the general idea of combinatorial sieve, since the more sophisticated sieves are extensions of its basic idea / Mestrado / Matematica Aplicada / Mestre em Matemática Aplicada

La thèse concerne différents aspects de la répartition des fonctions arithmétiques.1. Deshouillers, Iwaniec et Luca se sont récemment intéressés à la répartition modulo 1 de suites qui sont des valeurs moyennes de fonctions multiplicatives, par exemple phi(n)/n où phi est la fonction d'Euler. Nous étendons leur travail à la densité modulo 1 de suites qui sont des valeurs moyennes sur des suites polynômiales, typiquement n^2+1.2. On sait depuis les travaux de Katai, il y a une quarantaine d'années que la fonction de répartition des valeurs de phi(p-1)/(p-1) (où p parcourt les nombres premiers) est continue, purement singulière, strictement croissante entre 0 et 1/2. On précise cette étude en montrant que cette fonction de répartition a une dérivée infinie à gauche de tout point phi(2n)/(2n). / Abstract