Master of Science / Department of Statistics / Wei-Wen Hsu / The zero-inflated Poisson (ZIP) model consists of a Poisson model and a degenerate distribution at zero. Under this model, zero counts are generated from two sources, representing a heterogeneity in the population. In practice, it is often interested to evaluate this heterogeneity is consistent with the observed data or not. Most of the existing methodologies to examine this heterogeneity are often assuming that the Poisson mean is a function of nuisance parameters which are simply the coefficients associated with covariates. However, these nuisance parameters can be misspecified when performing these methodologies. As a result, the validity and the power of the test may be affected. Such impact of misspecification has not been discussed in the literature. This report primarily focuses on investigating the impact of misspecification on the performance of score test for homogeneity in ZIP models. Through an intensive simulation study, we find that: 1) under misspecification, the limiting distribution of the score test statistic under the null no longer follows a chi-squared distribution. A parametric bootstrap methodology is suggested to use to find the true null limiting distribution of the score test statistic; 2) the power of the test decreases as the number of covariates in the Poisson mean increases. The test with a constant Poisson mean has the highest power, even compared to the test with a well-specified mean. At last, simulation results are applied to the Wuhan Inpatient Care Insurance data which contain excess zeros.

This thesis deals with the goodness of fit tests in nonparametric model in the presence of unknown parameters of the probability distribution. The first part is devoted to understanding of the theoretical basis. We compare two methodologies for the construction of test statistics with application of empirical characteristic and empirical distribution functions. We use kernel estimates of regression functions and parametric bootstrap method to approximate the critical values of the tests. In the second part of the thesis, the work is complemented with the simulation study for different choices of weighting functions and parameters. Finally we illustrate the use and the comparison of goodness of fit tests on the example with the real data set. Powered by TCPDF (www.tcpdf.org)

Partant du postulat selon lequel la principale justification de la répression de formes d’expressions réside dans leur nocivité supposée, nous tentons de répondre à la question suivante : comment définir des limites claires et cohérentes à la liberté d’expression ? L’analyse des controverses relatives à la pornographie, et en particulier de la manière dont les notions de liberté d’expression et de nuisance ont été articulées, contribue à répondre à cette question générale. À travers l’analyse des débats portant sur la restriction des représentations sexuelles, nous tentons de montrer que les parties en présence ne sont pas parvenues à définir la notion de « nuisance » de manière claire et satisfaisante, et ne permettent pas, dès lors, de définir avec précision les limites légitimes de la liberté d’expression. Les deux voies théoriques alternatives que nous avons identifiées, les conceptions instrumentales et déontologiques de la liberté d’expression, ne se révèlent pas plus convaincantes. Nous montrons néanmoins qu’il est possible de préciser le principe de non-nuisance en y intégrant deux éléments auparavant négligés : la subjectivité du récepteur, et les rapports d’autorité qui existent entre le locuteur et le récepteur. Nous défendons ainsi l’idée que le principe de non-nuisance reste l’instrument le plus clair et le plus cohérent pour fonder les limites de la liberté d’expression, à condition de l’amender et de le compléter. / Acknowledging the fact that the main justification to restrict some forms of expression lies in the harm they may cause to others, this thesis aims at answering the following question: how do we define clear and coherent limits of the freedom of expression? The study of the controversies regarding pornography, and particularly the way in which the concepts of freedom of expression and harm are closely linked together, is an important contribution in order to answer this vast subject. Through the analysis of debates with regard to sexual representations, this thesis aims at gaining a deeper understanding on how the authors were unsuccessful in defining the notion of « harm » in a clear and convincing way, and fail at allowing to set precisely the legitimate limits of freedom of expression. The two alternative theoretical approaches that were identified and established - the instrumental and deontological conceptions of freedom of expression – were not proven to be more satisfactory either. However, this research confirms that the harm principle can be clarified if two previously neglected aspects are included in the analysis: the receiver’s subjectivity, and the authority relationship established between the speaker and the viewer. Thus, it is argued that the harm principle, given that it is modified and completed, remains the most effective and adequate tool in order to ground the limits of freedom of expression.

Conflicts between humans and black bears (Ursus americanus) jeopardize the safety of both humans and bears, especially when bears become food-conditioned to anthropogenic food sources in areas such as campgrounds. Interest in using non-lethal techniques, such as aversive conditioning, to manage such conflicts is growing. I conducted a captive experiment at The Wildlife Science Center in Minnesota and two field experiments in the La Sal Mountains, Utah, to investigate the effects of taste aversion conditioning using thiabendazole (TBZ) with a novel flavor cue and food removal on black bear food consumption and visitation to human food sources. In 2007, I conducted food trials with 6 captive black bears (3 control, 3 treatment). Controls received 1 kg baked goods scented with a peppermint-canola oil mixture and treatments received 1 kg baked goods also scented with a peppermint-canola oil mixture but mixed with 10-20 g TBZ. In the 2007 field experiment, I baited 24 field sites with 300 g of baked goods during a baseline phase for approximately 3 weeks. Half of these sites were then treated with 10 g of TBZ and camphor during a treatment phase for 4 weeks. In 2008, I baited 22 sites with 300 g of baked goods during a baseline phase for approximately 4 weeks. I then removed food and discontinued baiting at half of the sites for 4 weeks. Infrared cameras and barbed-wire hair snags were established at field sites to document bear visitation. I did not establish taste aversion in treated bears in captivity and bears fully consumed food in the majority of trials. Treating food supplies with 10 g TBZ and camphor flavor did not significantly reduce bear visitation (P = 0.615) or food consumption at field sites (P = 0.58). However, I observed a significant reduction in bear activity at sites where food was removed (P = 0.006). Potential reasons for my failure to reduce bear visitation using thiabendazole include insufficient conditioning, reluctance of bears to desist in investigating sites that previously contained untreated food, and masking of a treatment effect due to continued encounters of sites by new individuals.

An interdisciplinary approach at the interface between geodesy and geophysics has recently resolved several Earth science problems at regional and global scales. I use the term “geophysical geodesy” to distinguish the technical and theoretical aspect of geodesy from geophysical applications of geodetic techniques. Using a wide range of Earth observation data, I study the spatio-temporal characteristics of Earth surface deformation in the United States associated with several geophysical processes, including natural and anthropogenic subsidence and uplift, regional relative sea-level rise, and continental hydrological loading. The theoretical portion of this dissertation applies loading theory and develops a new hybrid method to improve the estimate of hydrologically-induced vertical deformation at time scales from sub-annual to multi-annual. The application part of this dissertation benefits from GPS and other geodetic and geologic data sets to study and model Earth’s surface uplift due to CO2 injection at an oil reservoir in coastal Texas, and coastal subsidence and nuisance flooding along the Mississippi River Delta and eastern seaboard of the United States.

Les principaux enjeux de la gestion des moustiques en France sont la prévention des risques vectoriels et le contrôle de la nuisance. A travers une étude historique sur la transmission du paludisme dans les vallées alpines, nous avons démontré que les aménagements hydrauliques devaient être considérés comme des facteurs à risque pour le développement de maladies vectorielles. L'estimation de la nuisance perçue permettrait d'adapter le contrôle des moustiques aux attentes de la population. L'expérience de deux enquêtes sociologiques a conduit à poser les bases d'un observatoire de la nuisance, qui emploierait conjointement et régulièrement les outils sociologiques et les échantillonnages sur le terrain, pour une démoustication raisonnée. Les moustiques du genre Coquillettidia représentent une nuisance mais sont inefficacement contrôlés, les larves vivant fixées aux racines des végétaux en profondeur, à l'abris des traitements. A l'issue d'une vaste étude de terrain, une liste d'indicateurs écologiques caractéristiques des gîtes larvaires à Coquillettidia a été dressée pour faciliter leur localisation. La nature des plantes aquatiques et la présence de kairomones (p-crésol, acides laurique et heptadécanoïque), le courant, la salinité, la turbidité, la pérennité de la pièce d'eau et son entretien, sont très probablement impliqués dans le choix des gîtes colonisés par ces moustiques. Des tests comportementaux couplés à des analyses chimiques en laboratoire ont permis de déterminer les mécanismes gouvernant la fixation des larves aux racines, dont dépend leur survie. Nous avons mis en évidence l'implication de substances allélochimiques généralistes (glycérol, uracile, thymine, uridine et thymidine) dans la reconnaissance des plantes hôtes par les larves. L'influence de l'intensité lumineuse, de la teneur en oxygène de l'eau, et de la morphologie des racines sur le comportement de fixation a été démontrée. Enfin une stratégie larvicide adaptée à l'écologie particulière de Coquillettidia a été mise au point. Nous avons prouvé que des plantes aquatiques pouvaient acheminer un insecticide systémique, le diméthoate, par voie phloémienne jusqu'aux racines, et libérer des composés toxiques au contact des larves. La fixation est fortement perturbée (- 85%), ce qui amplifie l'efficacité de cette stratégie.

Pouvons-nous donner un indice fiable de l'exposition d'un individu à des sources multiples (Bruit + Odeur) ? Ce travail propose une approche théorique et analytique de la multi-exposition in situ, en collaboration avec le GRECA (Groupe de Recherche sur l'Environnement en Chimie Appliquée). Elle s'inscrit dans une démarche pluridisciplinaire, au croisement des sciences pour l'ingénieur et de la compréhension des phénomènes urbains.<br />Le cas d'étude mené ici est celui du complexe chimique Rhôdia du Pont-de-Claix (38), situé aujourd'hui au coeur du tissu urbain. Ce phénomène de multi-exposition fait appel à plusieurs champs de savoirs et de compétences, dont témoignent ses différentes dimensions constitutives que nous nous proposons de passer en revue : dimensions techniques, dimensions sensorielles, dimensions sociales, et dimensions spatiales et urbaines.<br />Les résultats de ce travail conduisent à rechercher un indice de l'exposition de l'individu dans son intentionnalité et son attentionnalité du signal perçu, comme autant de consensus factices résultant d'un monde physique perçu et d'un monde sensible vécu et idéalisé. Techniquement, une méthode d'analyse de la qualité de l'air (BTEX et Aldéhydes) spécifique au in situ, a été mis au point avec le GRECA. D'autre part, une méthode d'enquête, basée sur la réactivation sensorielle (à partir des travaux précurseurs sur l'écoute réactivée de Augoyard), a été développée pour les besoins particuliers du sujet.

En aquesta tesi s'estudia el problema d'estimació cega de segon ordre en comunicacions digitals. En aquest camp, els símbols transmesos esdevenen paràmetres no desitjats (nuisance parameters) d'estadística no gaussiana que degraden les prestacions de l'estimador. En aquest context, l'estimador de màxima versemblança (ML) és normalment desconegut excepte si la relació senyal-soroll (SNR) és prou baixa. En aquest cas particular, l'estimador ML és una funció quadràtica del vector de dades rebudes o, equivalentment, una transformació lineal de la matriu de covariància mostral. Aquesta característica es compartida per altres estimadors importants basats en el principi de màxima versemblança com ara l'estimador ML gaussià (GML) i l'estimador ML condicional (CML). Així mateix, l'estimador MUSIC, i altres mètodes de subespai relacionats amb ell, es basen en la diagonalització de la matriu de covariància mostral. En aquest marc, l'aportació principal d'aquesta tesi és la deducció i avaluació de l'estimador òptim de segon ordre per qualsevol SNR i qualsevol distribució dels nuisance parameters.El disseny d'estimadors quadràtics en llaç obert i llaç tancat s'ha plantejat de forma unificada. Pel que fa als estimadors en llaç obert, s'han derivat els estimadors de mínim error quadràtic mig i mínima variància considerant que els paràmetres d'interès són variables aleatòries amb una distribució estadística coneguda a priori però, altrament, arbitrària. A partir d'aquest plantejament Bayesià, els estimadors en llaç tancat es poden obtenir suposant que la distribució a priori dels paràmetres és altament informativa. En aquest model de petit error, el millor estimador quadràtic no esbiaixat, anomenat BQUE, s'ha formulat sense convenir cap estadística particular pels nuisance parameters. Afegit a això, l'anàlisi de l'estimador BQUE ha permès calcular quina és la fita inferior que no pot millorar cap estimador cec que utilitzi la matriu de covariància mostral. Probablement, el resultat principal de la tesi és la demostració de què els estimadors quadràtics són capaços d'utilitzar la informació estadística de quart ordre dels nuisance parameters. Més en concret, s'ha demostrat que tota la informació no gaussiana de les dades que els mètodes de segon ordre són capaços d'aprofitar apareix reflectida en els cumulants de quart ordre dels nuisance parameters. De fet, aquesta informació de quart ordre esdevé rellevant si el mòdul dels nuisance parameters és constant i la SNR és moderada o alta. En aquestes condicions, es demostra que la suposició gaussiana dels nuisance parameters dóna lloc a estimadors quadràtics no eficients. Un altre resultat original que es presenta en aquesta memòria és la deducció del filtre de Kalman estès de segon ordre, anomenat QEKF. L'estudi del QEKF assenyala que els algoritmes de seguiment (trackers) de segon ordre poden millorar simultàniament les seves prestacions d'adquisició i seguiment si la informació estadística de quart ordre dels nuisance parameters es té en compte. Una vegada més, aquesta millora és significativa si els nuisance parameters tenen mòdul constant i la SNR és prou alta. Finalment, la teoria dels estimadors quadràtics plantejada s'ha aplicat en alguns problemes d'estimació clàssics en l'àmbit de les comunicacions digitals com ara la sincronització digital no assistida per dades, el problema de l'estimació del temps d'arribada en entorns amb propagació multicamí, la identificació cega de la resposta impulsional del canal i, per últim, l'estimació de l'angle d'arribada en sistemes de comunicacions mòbils amb múltiples antenes. Per cadascuna d'aquestes aplicacions, s'ha realitzat un anàlisi intensiu, tant numèric com asimptòtic, de les prestacions que es poden aconseguir amb mètodes d'estimació de segon ordre. / This thesis deals with the problem of blind second-order estimation in digital communications. In this field, the transmitted symbols appear as non-Gaussian nuisance parameters degrading the estimator performance. In this context, the Maximum Likelihood (ML) estimator is generally unknown unless the signal-to-noise (SNR) is very low. In this particular case, if the SNR is asymptotically low, the ML solution is quadratic in the received data or, equivalently, linear in the sample covariance matrix. This significant feature is shared by other important ML-based estimators such as, for example, the Gaussian and Conditional ML estimators. Likewise, MUSIC and other related subspace methods are based on the eigendecomposition of the sample covariance matrix. From this background, the main contribution of this thesis is the deduction and evaluation of the optimal second-order parameter estimator for any SNR and any distribution of the nuisance parameters.A unified framework is provided for the design of open- and closed-loop second-order estimators. In the first case, the minimum mean square error and minimum variance second-order estimators are deduced considering that the wanted parameters are random variables of known but arbitrary prior distribution. From this Bayesian approach, closed-loop estimators are derived by imposing an asymptotically informative prior. In this small-error scenario, the best quadratic unbiased estimator (BQUE) is obtained without adopting any assumption about the statistics of the nuisance parameters. In addition, the BQUE analysis yields the lower bound on the performance of any blind estimator based on the sample covariance matrix.Probably, the main result in this thesis is the proof that quadratic estimators are able to exploit the fourth-order statistical information about the nuisance parameters. Specifically, the nuisance parameters fourth-order cumulants are shown to provide all the non-Gaussian information that is utilizable for second-order estimation. This fourth-order information becomes relevant in case of constant modulus nuisance parameters and medium-to-high SNRs. In this situation, the Gaussian assumption is proved to yield inefficient second-order estimates.Another original result in this thesis is the deduction of the quadratic extended Kalman filter (QEKF). The QEKF study concludes that second-order trackers can improve simultaneously the acquisition and steady-state performance if the fourth-order statistical information about the nuisance parameters is taken into account. Once again, this improvement is significant in case of constant modulus nuisance parameters and medium-to-high SNRs.Finally, the proposed second-order estimation theory is applied to some classical estimation problems in the field of digital communications such as non-data-aided digital synchronization, the related problem of time-of-arrival estimation in multipath channels, blind channel impulse response identification, and direction-of-arrival estimation in mobile multi-antenna communication systems. In these applications, an intensive asymptotic and numerical analysis is carried out in order to evaluate the ultimate limits of second-order estimation.

In this dissertation, the generalized score tests based on weighted estimating equations are proposed for missing covariate data. Their properties, including the effects of nuisance functions on the forms of the test statistics and efficiency of the tests, are investigated. Different versions of the test statistic are properly defined for various parametric and semiparametric settings. Their asymptotic distributions are also derived. It is shown that when models for the nuisance functions are correct, appropriate test statistics can be obtained via plugging the estimates of the nuisance functions into the appropriate test statistic for the case that the nuisance functions are known. Furthermore, the optimal test is obtained using the relative efficiency measure. As an application of the proposed tests, a formal model validation procedure is developed for generalized linear models in the presence of missing covariates. The asymptotic distribution of the data driven methods is provided. A simulation study in both linear and logistic regressions illustrates the applicability and the finite sample performance of the methodology. Our methods are also employed to analyze a coronary artery disease diagnostic dataset.

The way a city sounds has something to do with how its residents move around and interact. It also has to do with the decisions made by the local government, as it tries to eliminate “harmful” and “unnecessary” sounds. This dissertation discusses how residents of São Paulo deal with noise conflicts and the sound-politics these mediations entail. The concept of sound-politics, which I develop throughout the dissertation, intersects with other concepts such as body politics and space politics and as such it denotes controversial sounds as instantiations of individual or collective differences. Here I focus on how state agents mediate controversial sounds. The narrative draws heavily on actor-network theory, which investigates social interaction as an open-ended and localized network made of humans and non-humans. The first part of the narrative discusses São Paulo’s spatial organization, noise legislation, and the enforcement of this legislation. Drawing from ethnography at meetings designed to review noise measurement standards, and at São Paulo’s anti-noise agency, I show the series of negotiations that accompany the identification and prosecution of noisemakers in the city. The second part of the dissertation looks closely at street parties that take place in the suburbs of São Paulo. Known as pancadões (“big thumps,” in reference to the loudness), these parties feature funk carioca and funk ostentação, two styles of popular music, among lower class teenagers in São Paulo’s suburbs. Based on fieldwork among youth, community meetings, and on interviews with police officers, I show how these parties have gone through noisification processes since they first emerged around 2008. / text