近年來由於地理資訊系統（GIS）的快速發展發，空間資料分析開始受到重視並在社會科學領域中逐漸扮演重要的角色。雖然一般的統計方法已在傳統資料分析上發展已久，然而它們卻不能有效地說明空間性資料，並且無法充分處理空間相依或空間異質性問題。一般而言，空間資料分析主要有兩個分派:模型導向學派與資料導向學派。本文研究目的在於應用空間統計方法合理且充分地評估房地產價值，研究方法包含地理統計(克利金和共克利金)、地理加權迴歸與空間特徵價格模型等，並且以台中市不動產資料進行實證探究。這項新的研究技術在不動產評價領域中將可提供更好的解析能力，使其在評價過程中或是不動產投資決策時，成為一個更強而有力的分析工具。 / In recent years, spatial data analysis has received significant awareness and played an important role in social science because of the rapid development of Geographic Information System (GIS). Although classic statistical methods are attractive in traditional data analysis, they cannot be executed seriously for spatial data. Standard statistical techniques didn’t sufficiently deal with spatial dependence or spatial heterogeneity issues. Generally, the model-driven method and the data-driven method are mainly the two branches of the spatial data analysis. The purpose of this paper is to apply spatial statistics methods including geostatistical methods (kriging and cokiging), geographically weighted regression, and spatial hedonic price models to real estate analysis. It seems to be completely reasonable and sufficient. The real estate data in Taichung city (Taiwan) is used to carry out our exploration. These techniques give better insight in the field of real estate assessment. They can apply a good instrument in mass appraisal and decision concerning real estate investment.

The research in the field of travel demand modeling is driven by the need to understand individuals’ behavior in the context of travel-related decisions as accurately as possible. In this regard, the activity-based approach to modeling travel demand has received substantial attention in the past decade, both in the research arena as well as in practice. At the same time, recent efforts have been focused on more fully realizing the potential of activity-based models by explicitly recognizing the multi-dimensional nature of activity-travel decisions. However, as more behavioral elements/dimensions are added, the dimensionality of the model systems tends to explode, making the estimation of such models all but infeasible using traditional inference methods. As a result, analysts and practitioners often trade-off between recognizing attributes that will make a model behaviorally more representative (from a theoretical viewpoint) and being able to estimate/implement a model (from a practical viewpoint). An alternative approach to deal with the estimation complications arising from multi-dimensional choice situations is the technique of composite marginal likelihood (CML). This is an estimation technique that is gaining substantial attention in the statistics field, though there has been relatively little coverage of this method in transportation and other fields. The CML approach is a conceptually and pedagogically simpler simulation-free procedure (relative to traditional approaches that employ simulation techniques), and has the advantage of reproducibility of the results. Under the usual regularity assumptions, the CML estimator is consistent, unbiased, and asymptotically normally distributed. The discussion above indicates that the CML approach has the potential to contribute in the area of travel demand modeling in a significant way. For example, the approach can be used to develop conceptually and behaviorally more appealing models to examine individuals’ travel decisions in a joint framework. The overarching goal of the current research work is to demonstrate the applicability of the CML approach in the area of activity-travel demand modeling and to highlight the enhanced features of the choice models estimated using the CML approach. The goal of the dissertation is achieved in three steps as follows: (1) by evaluating the performance of the CML approach in multivariate situations, (2) by developing multidimensional choice models using the CML approach, and (3) by demonstrating applications of the multidimensional choice models developed in the current dissertation. / text

No description available.

[pt] O processo de ocupação humana no território brasileiro tem provocado, há séculos, profundas alterações no meio ambiente. Desde os primeiros habitantes, a natureza tem sofrido alguma mudança no seu formato, em algum grau de escala. Se os povos tradicionais nativos possuem uma preocupação em conciliar a relação homem-meio ambiente, a invasão europeia inverteu toda essa lógica. Dos processos primitivos até o uso da mecanização moderna, o Brasil pós-1500 mostrou a força humana sobre a natureza e seus consequentes impactos. Apesar dos esforços maiores ou menores, das denúncias propaladas desde do período colonial, avança-se a destruição de ecossistemas de suma importância local e global. Diante desse cenário, o país possui biomas com enorme perigo de perdas irreversíveis se nada for feito. Neste contexto, esta tese tem como objetivo entender essas transformações com olhares histórico, geográfico e econômico, culminando numa análise quantitativa do uso da terra de 2004 a 2019 focada nos municípios dos dois principais biomas brasileiros ameaçados, Cerrado e Amazônia. Por meio de diferentes instrumentos da econometria espacial cujos princípios são as primeiras leis da Geografia, se pretende identificar os fatores que modificaram o uso da terra no período citado e propor medidas para atenuar o impacto do principal vetor de desmatamento direto nos biomas citados: a pecuária. Por meio dessa análise, pretende-se avaliar como a intensificação da bovinocultura de corte altera as emissões, a configuração do uso da terra, pode auxiliar no cumprimento da Lei de Proteção da Vegetação Nativa e gerar ganhos econômicos dentro de uma área com mais de 1.700 municípios. O objetivo final é determinar qual a melhor política de intensificação da atividade de pecuária de corte considerando o trade-off meio ambiente e economia. A escolha permitirá que o país consiga oferecer aos consumidores domésticos e internacionais uma atividade mais sustentável com minimização do impacto ambiental gerado e receitas majoradas. Com isso, a tese mostra que as alterações no uso da terra em quatro diferentes classes – agricultura, pecuária, floresta e outros usos – são fenômenos locais cujo somatório tem um impacto global relevante. Os resultados reforçam a importância de políticas públicas ambientais e agrícolas e observarem as diferenças entre os municípios em sua formulação e implementação. Há, para o Brasil, um imenso potencial tanto de ampliar a área florestal com a intensificação e de restaurar grandes áreas liberadas com o incremento do sistema produtivo de carne bovina. Assim, o país tem nas suas mãos uma chance ímpar de ter uma pecuária sustentável econômica e ambientalmente com uma política governamental que considere a diversidade local. / [en] The process of human occupation in Brazilian territory has caused profound changes in the environment for centuries. Since the first inhabitants, nature has undergone some change in its shape, to some degree of scale. If native traditional peoples have a concern to reconcile the man-environment relationship, the European invasion has reversed all this logic. From primitive processes to the use of modern mechanization, post-1500 Brazil showed human strength under nature and its consequent impacts. Despite the major or minor efforts, the denunciations proposed since the colonial period, the destruction of ecosystems of great local and global importance is advanced. Given this scenario, the country has biomes with enormous danger of irreversible losses if nothing is done. In this context, this thesis aims to understand these transformations in historical, geographical, and economic views, culminating in a quantitative analysis of land use from 2004 to 2019 focused on the municipalities of the two main threatened Brazilian biomes, Cerrado and Amazon. Through different instruments of spatial econometrics whose principles are the first laws of geography, it is intended to identify the factors that modified land use in the period mentioned and propose measures to mitigate the impact of the main vector of direct deforestation in the biomes mentioned: livestock. Through this analysis, it is intended to evaluate how the intensification of beef cattle culture alters emissions, the configuration of land use, can assist in complying with the Native Vegetation Protection Law and generates economic gains within an area with more than 1,700 municipalities. The goal is to determine the best policy of intensification of beef livestock activity considering the trade-off environment and economy. The choice will allow the country to be able to offer domestic and international consumers a more sustainable activity with minimization of the environmental impact generated and increased revenues. Considering all these facts, the thesis shows that changes in land use in four different classes – agriculture, livestock, forest, and other uses – are local phenomena whose sum has a relevant global impact. The results reinforce the importance of environmental and agricultural public policies and the observer in the differences between the municipalities in their formulation and implementation. For Brazil, there is an immense potential for both expanding the forest area with intensification and restoring large areas released with the increase of the beef production system. The country has in its hands a unique chance to have a sustainable livestock economically and environmentally with a government policy that considers local diversity. / [fr] Le processus d occupation humaine sur le territoire brésilien a provoqué de profonds changements dans l environnement pendant des siècles. Depuis les premiers habitants, la nature a subi quelques changements dans sa forme, à un certain degré d échelle. Si les peuples autochtones traditionnels ont le souci de concilier la relation hommeenvironnement, l invasion européenne a inversé toute cette logique. Des processus primitifs à l utilisation de la mécanisation moderne, le Brésil post-1500 a montré la force humaine sous la nature et ses impacts conséquents. Malgré les efforts majeurs ou mineurs, les dénonciations proposées depuis la période coloniale, la destruction d écosystèmes d une grande importance locale et mondiale progresse. Compte tenu de ce scénario, le pays a des biomes avec un énorme danger de pertes irréversibles si rien n est fait. Dans ce contexte, cette thèse vise à comprendre ces transformations avec des vues historiques, géographiques et économiques aboutissant à une analyse quantitative de l utilisation des terres de 2004 à 2019 des municipalités des deux principaux biomes brésiliens menacés, le Cerrado et l Amazonie. A travers de différents instruments d économétrie spatiale dont les principes sont les premières lois de la géographie, il est prévu d identifier les facteurs qui ont modifié l utilisation des terres dans la période mentionnée et de proposer des mesures pour atténuer l impact du principal vecteur de déforestation directe dans les biomes mentionnés : l élevage. Grâce à cette analyse, il est prévu d évaluer comment l intensification de la culture des bovins modifie les émissions, la configuration de l utilisation des terres, peut aider à se conformer à la Loi sur la protection de la végétation native et générer des gains économiques dans une zone comptant plus de 1.700 municipalités. L objectif est de déterminer la meilleure politique d intensification de l activité du bétail en tenant compte de l environnement et de l économie compromis. Ce choix permettra au pays d être en mesure d offrir aux consommateurs domestiques et internationaux une activité plus durable avec une minimisation de l impact environnemental généré et une augmentation des revenus. Avec cela, la thèse montre que les changements dans l utilisation des terres dans quatre classes différentes – l agriculture, l élevage, la forêt et d autres utilisations – sont des phénomènes locaux dont la somme a un impact mondial pertinent. Les résultats renforcent l importance des politiques publiques environnementales et agricoles et regardent les différences entre les municipalités. Pour le Brésil, il existe un immense potentiel à la fois pour l expansion de la zone forestière avec l intensification et la restauration de grandes zones libérées avec l augmentation du système de production de bœuf. Le pays a entre ses mains une chance unique d avoir un élevage durable économique et écologiquement avec une politique gouvernementale qui prend en compte la diversité locale.

No description available.

Submitted by Paulo Gustavo Grahl (pgrahl@fgvmail.br) on 2013-10-18T05:32:44Z No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) / Approved for entry into archive by ÁUREA CORRÊA DA FONSECA CORRÊA DA FONSECA (aurea.fonseca@fgv.br) on 2013-10-28T18:22:53Z (GMT) No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2013-10-29T18:24:15Z (GMT) No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) / Made available in DSpace on 2013-10-29T18:25:35Z (GMT). No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) Previous issue date: 2012-12-22 / Esta dissertação concentra-se nos processos estocásticos espaciais definidos em um reticulado, os chamados modelos do tipo Cliff & Ord. Minha contribuição nesta tese consiste em utilizar aproximações de Edgeworth e saddlepoint para investigar as propriedades em amostras finitas do teste para detectar a presença de dependência espacial em modelos SAR (autoregressivo espacial), e propor uma nova classe de modelos econométricos espaciais na qual os parâmetros que afetam a estrutura da média são distintos dos parâmetros presentes na estrutura da variância do processo. Isto permite uma interpretação mais clara dos parâmetros do modelo, além de generalizar uma proposta de taxonomia feita por Anselin (2003). Eu proponho um estimador para os parâmetros do modelo e derivo a distribuição assintótica do estimador. O modelo sugerido na dissertação fornece uma interpretação interessante ao modelo SARAR, bastante comum na literatura. A investigação das propriedades em amostras finitas dos testes expande com relação a literatura permitindo que a matriz de vizinhança do processo espacial seja uma função não-linear do parâmetro de dependência espacial. A utilização de aproximações ao invés de simulações (mais comum na literatura), permite uma maneira fácil de comparar as propriedades dos testes com diferentes matrizes de vizinhança e corrigir o tamanho ao comparar a potência dos testes. Eu obtenho teste invariante ótimo que é também localmente uniformemente mais potente (LUMPI). Construo o envelope de potência para o teste LUMPI e mostro que ele é virtualmente UMP, pois a potência do teste está muito próxima ao envelope (considerando as estruturas espaciais definidas na dissertação). Eu sugiro um procedimento prático para construir um teste que tem boa potência em uma gama de situações onde talvez o teste LUMPI não tenha boas propriedades. Eu concluo que a potência do teste aumenta com o tamanho da amostra e com o parâmetro de dependência espacial (o que está de acordo com a literatura). Entretanto, disputo a visão consensual que a potência do teste diminui a medida que a matriz de vizinhança fica mais densa. Isto reflete um erro de medida comum na literatura, pois a distância estatística entre a hipótese nula e a alternativa varia muito com a estrutura da matriz. Fazendo a correção, concluo que a potência do teste aumenta com a distância da alternativa à nula, como esperado. / This dissertation focus on spatial stochastic process on a lattice (Cliff & Ord--type of models). My contribution consists of using Edgeworth and saddlepoint series to investigate small sample size and power properties of tests for detecting spatial dependence in spatial autoregressive (SAR) stochastic processes, and proposing a new class of spatial econometric models where the spatial dependence parameters that enter the mean structure are different from the ones in the covariance structure. This allows a clearer interpretation of models' parameters and generalizes the set of local and global models suggested by Anselin (2003) as an alternative to the traditional Cliff & Ord models. I propose an estimation procedure for the model's parameters and derive the asymptotic distribution of the parameters' estimators. The suggested model provides some insights on the structure of the commonly used mixed regressive, spatial autoregressive model with spatial autoregressive disturbances (SARAR). The study of the small sample properties of tests to detect spatial dependence expands on the existing literature by allowing the neighborhood structure to be a nonlinear function of the spatial dependence parameter. The use of series approximations instead of the often used Monte Carlo simulation allows a simple way to compare test properties across different neighborhood structures and to correct for size when comparing power. I obtain the power envelope for testing the presence of spatial dependence in the SAR process using the optimal invariant test statistic, which is also locally uniformly most powerful invariant (LUMPI). I have found that the LUMPI test is virtually UMP since its power is very close to the power envelope. I suggest a practical procedure to build a test that, while not UMP, retain good power properties in a wider range for the spatial parameter when compared to the LUMPI test. I find that power increases with sample size and with the spatial dependence parameter -- which agrees with the literature. However, I call into question the consensus view that power decreases as the spatial weight matrix becomes more densely connected. This finding in the literature reflects an error of measure because the hypothesis being compared are at very different statistical distance from the null. After adjusting for this, the power is larger for alternative hypothesis further away from the null -- as one would expect.

Ce mémoire de thèse porte sur la statistique inférentielle des données spatiales et/ou fonctionnelles. En effet, nous nous sommes intéressés à l’estimation de paramètres inconnus de certains modèles à partir d’échantillons obtenus par un processus d’échantillonnage aléatoire ou non (stratifié), composés de variables indépendantes ou spatialement dépendantes.La spécificité des méthodes proposées réside dans le fait qu’elles tiennent compte de la nature de l’échantillon étudié (échantillon stratifié ou composé de données spatiales dépendantes).Tout d’abord, nous étudions des données à valeurs dans un espace de dimension infinie ou dites ”données fonctionnelles”. Dans un premier temps, nous étudions les modèles de choix binaires fonctionnels dans un contexte d’échantillonnage par stratification endogène (échantillonnage Cas-Témoin ou échantillonnage basé sur le choix). La spécificité de cette étude réside sur le fait que la méthode proposée prend en considération le schéma d’échantillonnage. Nous décrivons une fonction de vraisemblance conditionnelle sous l’échantillonnage considérée et une stratégie de réduction de dimension afin d’introduire une estimation du modèle par vraisemblance conditionnelle. Nous étudions les propriétés asymptotiques des estimateurs proposées ainsi que leurs applications à des données simulées et réelles. Nous nous sommes ensuite intéressés à un modèle linéaire fonctionnel spatial auto-régressif. La particularité du modèle réside dans la nature fonctionnelle de la variable explicative et la structure de la dépendance spatiale des variables de l’échantillon considéré. La procédure d’estimation que nous proposons consiste à réduire la dimension infinie de la variable explicative fonctionnelle et à maximiser une quasi-vraisemblance associée au modèle. Nous établissons la consistance, la normalité asymptotique et les performances numériques des estimateurs proposés.Dans la deuxième partie du mémoire, nous abordons des problèmes de régression et prédiction de variables dépendantes à valeurs réelles. Nous commençons par généraliser la méthode de k-plus proches voisins (k-nearest neighbors; k-NN) afin de prédire un processus spatial en des sites non-observés, en présence de co-variables spatiaux. La spécificité du prédicteur proposé est qu’il tient compte d’une hétérogénéité au niveau de la co-variable utilisée. Nous établissons la convergence presque complète avec vitesse du prédicteur et donnons des résultats numériques à l’aide de données simulées et environnementales.Nous généralisons ensuite le modèle probit partiellement linéaire pour données indépendantes à des données spatiales. Nous utilisons un processus spatial linéaire pour modéliser les perturbations du processus considéré, permettant ainsi plus de flexibilité et d’englober plusieurs types de dépendances spatiales. Nous proposons une approche d’estimation semi paramétrique basée sur une vraisemblance pondérée et la méthode des moments généralisées et en étudions les propriétés asymptotiques et performances numériques. Une étude sur la détection des facteurs de risque de cancer VADS (voies aéro-digestives supérieures)dans la région Nord de France à l’aide de modèles spatiaux à choix binaire termine notre contribution. / This thesis is about statistical inference for spatial and/or functional data. Indeed, weare interested in estimation of unknown parameters of some models from random or nonrandom(stratified) samples composed of independent or spatially dependent variables.The specificity of the proposed methods lies in the fact that they take into considerationthe considered sample nature (stratified or spatial sample).We begin by studying data valued in a space of infinite dimension or so-called ”functionaldata”. First, we study a functional binary choice model explored in a case-controlor choice-based sample design context. The specificity of this study is that the proposedmethod takes into account the sampling scheme. We describe a conditional likelihoodfunction under the sampling distribution and a reduction of dimension strategy to definea feasible conditional maximum likelihood estimator of the model. Asymptotic propertiesof the proposed estimates as well as their application to simulated and real data are given.Secondly, we explore a functional linear autoregressive spatial model whose particularityis on the functional nature of the explanatory variable and the structure of the spatialdependence. The estimation procedure consists of reducing the infinite dimension of thefunctional variable and maximizing a quasi-likelihood function. We establish the consistencyand asymptotic normality of the estimator. The usefulness of the methodology isillustrated via simulations and an application to some real data.In the second part of the thesis, we address some estimation and prediction problemsof real random spatial variables. We start by generalizing the k-nearest neighbors method,namely k-NN, to predict a spatial process at non-observed locations using some covariates.The specificity of the proposed k-NN predictor lies in the fact that it is flexible and allowsa number of heterogeneity in the covariate. We establish the almost complete convergencewith rates of the spatial predictor whose performance is ensured by an application oversimulated and environmental data. In addition, we generalize the partially linear probitmodel of independent data to the spatial case. We use a linear process for disturbancesallowing various spatial dependencies and propose a semiparametric estimation approachbased on weighted likelihood and generalized method of moments methods. We establishthe consistency and asymptotic distribution of the proposed estimators and investigate thefinite sample performance of the estimators on simulated data. We end by an applicationof spatial binary choice models to identify UADT (Upper aerodigestive tract) cancer riskfactors in the north region of France which displays the highest rates of such cancerincidence and mortality of the country.