abstract: The dynamics of urban water use are characterized by spatial and temporal variability that is influenced by associated factors at different scales. Thus it is important to capture the relationship between urban water use and its determinants in a spatio-temporal framework in order to enhance understanding and management of urban water demand. This dissertation aims to contribute to understanding the spatio-temporal relationships between single-family residential (SFR) water use and its determinants in a desert city. The dissertation has three distinct papers to support this goal. In the first paper, I demonstrate that aggregated scale data can be reliably used to study the relationship between SFR water use and its determinants without leading to significant ecological fallacy. The usability of aggregated scale data facilitates scientific inquiry about SFR water use with more available aggregated scale data. The second paper advances understanding of the relationship between SFR water use and its associated factors by accounting for the spatial and temporal dependence in a panel data setting. The third paper of this dissertation studies the historical contingency, spatial heterogeneity, and spatial connectivity in the relationship of SFR water use and its determinants by comparing three different regression models. This dissertation demonstrates the importance and necessity of incorporating spatio-temporal components, such as scale, dependence, and heterogeneity, into SFR water use research. Spatial statistical models should be used to understand the effects of associated factors on water use and test the effectiveness of certain management policies since spatial effects probably will significantly influence the estimates if only non-spatial statistical models are used. Urban water demand management should pay attention to the spatial heterogeneity in predicting the future water demand to achieve more accurate estimates, and spatial statistical models provide a promising method to do this job. / Dissertation/Thesis / Ph.D. Sustainability 2013

Submitted by Liliane Ferreira (ljuvencia30@gmail.com) on 2018-09-03T12:14:16Z No. of bitstreams: 2 Dissertação - Juliana Silva Prado - 2018.pdf: 3426266 bytes, checksum: 8677cc417b12183a4ded6032cee7d911 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2018-09-04T11:37:20Z (GMT) No. of bitstreams: 2 Dissertação - Juliana Silva Prado - 2018.pdf: 3426266 bytes, checksum: 8677cc417b12183a4ded6032cee7d911 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-09-04T11:37:20Z (GMT). No. of bitstreams: 2 Dissertação - Juliana Silva Prado - 2018.pdf: 3426266 bytes, checksum: 8677cc417b12183a4ded6032cee7d911 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2018-02-14 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Understanding what factors are behind human morphological variation has for many years been one of the key objectives of various research fields, namely evolutionary, genetic and anthropological biology. The morphological diversity of the human skull sparks great scientific interest, seeing as though quantitative data (due to the genetic complexity in play) showing the patterns of microevolution is useful for analyzing and understanding matters concerning the evolutionary history of populations, such as dispersal, gene flow, isolation by distance, large-scale expansion, among others. For this purpose, the use of multivariate techniques, such as Principal Component Analysis (PCA), has been supported to assess the human genetic variation on continents. Within this context, the key objective of this article was to characterize human cranial variation, utilizing PCA and Multivariate Spatial Correlation (MSC), so as to assess and identify possible evolutionary processes that contributed to the variation observed. To this end, cranial measurements available on the database obtained by W. Howells (57 variables), sourced from 1248 adult male specimens distributed throughout 30 locations (populations) in the world, were utilized. The results show that there has been spatial structuration of data, as indicated by the spatial autocorrelation statistics (Mantel Test 0.4077, P = 0.001; 59,64% of Moran's Index value with 0.05 significance and average correlogram with positive values in the first few distance bands and negative values in the subsequent bands). The use of PCA and MSC demonstrated that MSC was able to best capture the spatial pattern of data, increasing variation percentages from 54,74% to 69,33% in the first two principal components, where the techniques showed that 26 variables relative to cranial size had positive correlations in these components. The mapping and multivariate regression analyses utilizing environmental data and average dispersion age showed that the variation in the cranial size of populations followed a pattern of increase in cranial size correlated with low temperatures and recent colonization. The results obtained are consistent with Bergmann's Rule, which may thus be applied to modern humans. / Compreender quais são os fatores que estão por trás da variação morfológica humana tem sido há muitos anos um dos principais objetivos de diversas áreas de pesquisa, destacando a biologia evolutiva, genética e antropologia. A diversidade morfológica do crânio humano desperta grande interesse científico, onde o uso de dados quantitativos (devido à complexidade genética que o influencia) demostrando a atuação de processos microevolutivos, é útil para analisar e buscar compreender questões relativas a história evolutiva das populações, como eventos de dispersão genética, fluxo gênico, isolamento por distância e expansão em grande escala, dentre outros. Para tal, o uso das técnicas multivariadas, como a Análise de Componentes Principais (PCA), tem sido defendido para se avaliar a variação genética humana em regiões continentais. Nesse contexto, o objetivo central deste trabalho foi caracterizar a variação craniana humana, utilizando a PCA e a técnica de Correlação Espacial Multivariada (MSC), a fim de avaliar e identificar possíveis processos evolutivos que contribuíram para a variação observada. Para tal propósito, foram utilizadas as características métricas cranianas disponíveis no banco de dados obtido por W. Howells (57 variáveis) proveniente de 1248 espécimes adultos do sexo masculino distribuídos em 30 localidades (populações) pelo mundo. Os resultados demonstraram que houve estruturação espacial dos dados, indicado pelas estatísticas de autocorrelação espacial (Teste de Mantel 0.4077, P = 0.001; 59,64% dos índices I de Moran significativos a 0.05, e correlograma médio com valores positivos nas primeiras classes de distância e negativos nas seguintes). O uso da PCA e do MSC demonstraram que a técnica do MSC capturou melhor o padrão espacial dos dados, aumentando os valores da percentagem de variação passando de 54,74% para 69,33% nos 2 primeiros componentes principais, onde as técnicas demonstraram que 26 variáveis correspondentes a tamanho neurocraniano, possuíam correlações positivas nos dois primeiros componentes principais. Os mapas sintéticos e as análises de regressão multivariada utilizando dados ambientais e idade média de dispersão demonstraram que a variação do tamanho do crânio nas populações seguiu um padrão de aumento do tamanho craniano correlacionado a temperaturas baixas e idade de colonização recente. Os resultados obtidos são condizentes com a Regra Ecogeográfica de Bergmann, que pode então ser aplicada a humanos modernos.

Submitted by Erika Demachki (erikademachki@gmail.com) on 2016-06-02T17:26:11Z No. of bitstreams: 2 Dissertação - Camila Fernanda Costa - 2013.pdf: 2572672 bytes, checksum: 54a6e93e72c746e4888a6f3583d4f3ac (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Erika Demachki (erikademachki@gmail.com) on 2016-06-02T17:49:46Z (GMT) No. of bitstreams: 2 Dissertação - Camila Fernanda Costa - 2013.pdf: 2572672 bytes, checksum: 54a6e93e72c746e4888a6f3583d4f3ac (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2016-06-02T17:49:46Z (GMT). No. of bitstreams: 2 Dissertação - Camila Fernanda Costa - 2013.pdf: 2572672 bytes, checksum: 54a6e93e72c746e4888a6f3583d4f3ac (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2013-03-23 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / Hancornia speciosa (Apocynaceae) is a fruit species which has a wide distribution in areas of Cerrado vegetation type restricted sense. Its flowers have a complex pollination mechanism and unique, co-adapted to pollination by moths and butterflies and their fruits are dispersed by mammals large and medium-sized. It has high economic potential and its use has been conducted in an exploratory way. To implement conservation programs, breeding and enabling the commercial use of species, knowledge about the genetic variability, the spatial genetic structure, the reproductive system and gene flow in natural populations are needed. In this context, in the present study was performed the molecular characterization of the genetic variability in three stages of individuals (adults, juveniles and seedlings) and the assessment of genetic structure spatial (adults and juveniles), of the system of crossing and of the gene flow via pollen in a subpopulation of Hancornia speciosa located in the State Park of Serra de Jaragua, Jaragua-GO. To this end, 113 adults and 100 juveniles were sampled and georeferenced in an area of approximately 2.5ha. Of the total number of adults, 20 trees were selected matrices to obtain seedlings and formation of families of open pollination. Genomic DNA was obtained from the leaves of all individuals (adults, juveniles and seedlings) and was amplified with the use of seven microsatellite loci for obtaining of genotypes. The analyzes of genetic diversity, of spatial structure, rates of cross-fertilization and distance of dispersal of pollen were obtained from these genotypes. The total number of alleles at seven loci evaluated was 125, with an average of 17.8 alleles per locus. For adults the mean allele was 15.8, for the juveniles was 13.5and the seedlings were 11alleles. The average total values of heterozigosidade expected (He) and observed (Ho) were equal to 0.750 and 0.698, respectively. In adults He = 0.750 and Ho =0.714, in juveniles He =0.744 and Ho =0.679 and in seedlings He = 0. 712 and Ho =0. 763. These values indicate that the subpopulation evaluated presents high levels of genetic diversity. The fixation index(f) waspositive and significantforthe generations ofadults(0.052, p <0.05)andjuvenile(0.087, p <0.05), indicating the existence of inbreeding in this subpopulation. The analysis of spatial autocorrelation evidenced that kinship is weakly related to the geographical distance in in bothstagesof lifeevaluated (adults: b= -0.00223, R2 = 0. 000514, p < 0.05 and juveniles: b: - 0.00440, R2 = 0.00148489; p < 0.001).Concomitantly the values of Sp were low and the size of neighborhoods (Nb) were high for the two generations. This result shows that there is no restriction of gene flow via seed and corroborates the hypothesis that the dispersal by animals have high potential to disperse the seeds over long distances.The rates of crossfertilization multilocus (tm= 1.000) and single locus (ts = 0.972 a 1.29) were high and significantly different from zero in all families. The difference in the rate of the crossfertilization multilocus and single locus combined for all families analyzed (tm-ts = 0. 077) was also positive and significant, suggesting that 7% of crossings that occur in this population are between related individuals. The correlation of selfing negative(rs=-0.999), indicates absence of selfing and the correlation of paternity(rp=0.107) not significantly different from zero (SD = 0.135) shows that this subpopulation no full siblings. Every subpopulation resulting outcrossing and individuals are evaluated relatives at least to the 15 level of half-brothers. The paternity analysis assigned pollen donors to 75.2% (64) seedlings, and 24 (37.5%) assignments at a confidence level of 85%, 30 (46.8%) assignments at 95% probability assignments and 10(15.6%) at a confidence level of 99%.This low assignment can be explained by sampling: due to the size of the total area of the population and distribution in aggregate, many individuals may not have been sampled. Moreover, the loci battery used in this analysis could not demonstrated the optimal values for the combined exclusion. Although the maximum distance of pollination has been of 292m and covers the entire area evaluated, most events of pollination (77%) occurred at distances less than 200m. The flowering in mass, the distribution in aggregate and the floral structure are the main responsible for predominance of events of cross-pollination at short distances. / Hancornia speciosa (Apocynaceae) é uma espécie frutífera que apresenta distribuição ampla em áreas de fitofisionomia de Cerrado sentido restrito. Suas flores possuem um mecanismo de polinização complexo e único, adaptado á polinização por mariposas e borboletas. Seus frutos são dispersos por mamíferos de grande e médio porte. Possui alto potencial econômico e sua utilização tem sido realizada de forma exploratória. Para implementar programas de conservação e melhoramento, conhecimentos a cerca da variabilidade genética, da estrutura genética espacial, do sistema reprodutivo e do fluxo gênico nas populações naturais são necessários. Nesse contexto, no presente estudo foi realizada a caracterização molecular da variabilidade genética em três estágios de indivíduos (adultos, juvenis e plântulas) e a avaliação da estrutura genética espacial (adultos e juvenis), do sistema de cruzamento e do fluxo gênico via pólen em uma subpopulação de Hancornia speciosa localizada na região do Parque Estadual da Serra de Jaraguá, Jaraguá-GO. Para tanto, 113 indivíduos adultos e 100 juvenis foram amostrados e georeferenciados em uma área de aproximadamente 2,5 ha. Do total de adultos, 20 árvores matrizes foram selecionadas para obtenção de plântulas e formação das famílias de polinização aberta. O DNA genômico foi obtido a partir das folhas de todos os indivíduos (adultos, juvenis e plântulas) e foi amplificado com o uso de sete locos microssatélites para obtenção dos genótipos. As análises de diversidade genética, de estrutura espacial, taxas de fecundação cruzada e distância de dispersão de pólen foram obtidas a partir desses genótipos. O número total de alelos nos sete locos avaliados foi de 125, com média de 17,8 alelos por loco. Para os indivíduos adultos a média de alelos foi de 15,8, para os juvenis foi de 13,5 e nas plântulas foi de 11 alelos. Os valores totais médios de heterozigosidade esperada (He) e observada (Ho) foram iguais a 0, 750 e 0, 698. Nos adultos He = 0, 750 e Ho =0, 714, nos juvenis He =0, 744 e Ho =0, 679 e nas plântulas He = 0, 712 e Ho =0, 763. Esses valores indicam que a subpopulação avaliada apresenta altos níveis de diversidade genética. O índice de fixação (f) foi positivo e significativo para as gerações de indivíduos adultos (0, 052; p < 0, 05) e juvenis (0, 087; p < 0, 05), indicando a existência de endogamia nessa subpopulação. A análise de autocorrelação espacial evidenciou que o parentesco está fracamente relacionado com a distância geográfica em ambas as gerações de indivíduos avaliadas (adultos: b= -0, 00223, R2 = 0, 000514, p < 0,05 e juvenis: b: - 0, 00440, R2 = 0, 00148489; p < 0, 001). Concomitantemente os valores de Sp foram baixos e o tamanho das vizinhanças (Nb) foram altos para as duas gerações. Esse resultado mostra que não existe restrição ao fluxo gênico via semente e corrobora com a hipótese de que a dispersão por animais tem alto potencial para dispersar as sementes a longas distâncias. As taxas de cruzamento multiloco (tm = 1, 000) e uniloco (ts =0, 972 a 1, 29) foram altas e significativamente diferentes de zero em todas as famílias. A diferença da taxa de fecundação cruzada multiloco e uniloco combinada para todas as famílias analisadas (tm-ts = 0, 077) também foi positiva e significativa, sugerindo que 7% dos cruzamentos que ocorrem nessa população são entre indivíduos aparentados. A correlação de autofecundação negativa (rs = -0,999) indica ausência de autofecundação e a correlação de paternidade (rp= 0,107) não significativamente diferente de zero (SD = 0,135) mostra que 13 nessa subpopulação não existem irmãos completos. Toda a subpopulação é resultante de fecundação cruzada e os indivíduos avaliados são parentes no mínimo ao nível de meiosirmãos. A análise de paternidade atribuiu doadores de pólen para 75,2 % (64) plântulas, sendo 24 (37,5 %) atribuições ao nível de confiança de 85%, 30 (46,8%) atribuições ao nível de 95% de probabilidade e 10 atribuições, 15,6% ao nível de confiança de 99%. Essa baixa atribuição pode ser explicada pela amostragem: devido ao tamanho da área total da população e a distribuição em agregado, muitos indivíduos podem não ter sido amostrados. Além disso, a bateria de locos usada nessa análise não apresentou os valores considerados ótimos de probabilidade de exclusão combinada. Embora a distância máxima de polinização tenha sido de 292m e abrange toda a área avaliada, a maior parte dos eventos de polinização (77%) ocorreu a distâncias inferiores a 200m. A floração em massa, a distribuição em agregado e a estrutura floral são os principais responsáveis pelo predomínio de eventos de polinização cruzada a curtas distâncias.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-07-04T20:23:16Z No. of bitstreams: 1 claudioluizdecarvalho.pdf: 1237703 bytes, checksum: 32187853fc58c41f6d31975a9a0c1b13 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-07-13T16:23:21Z (GMT) No. of bitstreams: 1 claudioluizdecarvalho.pdf: 1237703 bytes, checksum: 32187853fc58c41f6d31975a9a0c1b13 (MD5) / Made available in DSpace on 2016-07-13T16:23:21Z (GMT). No. of bitstreams: 1 claudioluizdecarvalho.pdf: 1237703 bytes, checksum: 32187853fc58c41f6d31975a9a0c1b13 (MD5) Previous issue date: 2012-04-23 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Resumo: A literatura de transporte raramente considera a influência da interação espacial e das redes sociais sobre os fluxos de passageiros. A presente dissertação possui como objetivo a estimação da demanda de transporte aéreo por meio de dados na forma de fluxos de passageiros entre pares origem-destino (O-D), utilizando como método de estimação o modelo gravitacional ampliado, incorporando efeitos fixos, espaciais e de rede. Em relação ao modelo gravitacional, sua especificação se mostrou adequada nas estimações mostrando que tanto o tamanho dos municípios quanto a distância que os separa são variáveis relevantes para explicar o fluxo de passageiros entre os mesmos. O que mostra que a ―massa‖ do município, o PIB ou a população, são os principais determinantes para o incremento de passageiros em rotas aéreas, e que a distância se mostra uma restrição relevante para este incremento. Além disso, as variáveis que captam a autocorrelação espacial no modelo, ainda que menos significativas, são relevantes para explicar o fluxo de passageiros. / Abstract: The transport demand literature has rarely considered the influence of geographical interation or social network on passenger flows. This work aims to estimate air transport demand using an extended gravity model, controlling for spatial and network effects. As to the Gravity Model, the extended specification has been proper and has demonstrated the relevance of economic masses (GDP and population) as well the distance to explain passengers flows between the cities. Although spatial autocorrelation is a little significant, it is still useful to explain air passengers flow demand.

Introduction: One of the major barriers to developing an accurate tuberculosis (TB) surveillance program for Florida is the design and implementation of a sampling system that will adequately monitor and predict varying sizes and characteristics of county-level vulnerable endemic sub-populations and their explanatory covariates (e.g., living or working in a residential care facility). The aim of this research study is to envision an endemic, tuberculosis-related web-based interface for use by public health officials in the State of Florida which includes generating essential information such as a real-time syndrome-based reporting to regulate automated and immediate 'Alerts' to public health officials, doctors, hospitals and local community in ArcGIS. This study demonstrates the capability of an autocorrelation, time series, epidemiological, interpolative, and vulnerable predictive ArcGIS model to target tuberculosis at the county-level in the state of Florida. Methodology: The data for constructing an autocorrelation, probabilistic paradigm was acquired from the Centers for Disease Control and Prevention [CDC] in Atlanta, Georgia. The full dataset contained two points in time, allowing estimation of a mixed binomial model that aided in predicting the probability of tuberculosis by county. The random effects term in the ArcGIS model was comprised of spatially structured and stochastic effects (i.e., spatially unstructured) terms. These terms substituted for covariates in the model. The assumption was that random effects term in the endemic, TB–related, explanative, county-level, risk model had a frequency distribution that was bell-shaped (i.e., normally/Gaussian distributed) with a mean of zero. Results: The results indicated the empirical estimate had a mean of 0.0197 with a Shapiro-Wilk normality probability of 0.0027. The mean in the model was not exactly zero, although the forecasts indicated 0.06, which was not significantly different from zero. It was noted that the frequency distribution deviated from a bell-shaped curve. This random effects term accounted for roughly 41% of the variability in the observed probability of TB by county and yielded an under dispersed binomial model. An eigenvector spatial filter description of the random effects term involved 5 of 18 total eigenvectors, which portrayed noticeable positive spatial autocorrelation. The decomposition algorithm also revealed 4 of 25 eigenvectors portraying noticeable negative spatial autocorrelation. These two spatial filter components accounted for, respectively, roughly 16% and 10% of the variability in the probability of TB by county. The spatially unstructured random effects component accounted for roughly 15% of this variability. The final model revealed that from 2015 to 2020, Duval, Orange, and Broward counties would require immediate intervention in order to prevent TB transmission. The model also revealed that from 2025 to 2040 Hillsborough and Palm Beach counties could become hyper-endemic without implementation of control strategies. Conclusion: An endemic, TB-related, ArcGIS, autocorrelation eigenanalyses forecast, paradigm may be employed by public health officials in Florida to target, vulnerable, county level ,populations A Precede-Proceed model-based reporting mechanism may help disseminate the ArcGIS model results and help regulate automated and immediate 'Alerts' to public health officials, doctors, hospitals and local community at the county-level. An ArcGIS, web-based, epidemiological tool for data entry and communication can also allow real-time , predictive, real-time mapping of any TB county outbreaks Precede –Proceed model may be employed by county-level public health officials in Florida to disseminate and prioritize county-level, TB model, epidemiological, information to their constituents. In so doing, factors regulating outbreaks of county-level TB may be accurately identified.

Development of housing construction in the context of regional population development in Czechia Abstract The main aim of the thesis is to analyze and describe the development of housing construction in the context of regional population development in the Czech Republic in the years 2000- 2019. Spatial autocorrelation (LISA) and Webb's graph, which is used in analyzes of regional population development, were used for data analysis. Areas with above-average and, on the contrary, below-average housing construction are observed in four five-year periods. These areas are mutually compared in terms of their population development. The main result of the thesis is the confirmation of the assumption that in most cases, areas with above-average housing construction are also characterized by above-average population growth and are located in the hinterland of large cities (especially Prague) as part of the suburbanization process. Keywords: population development, housing construction, Webb's graph, spatial autocorrelation, Czechia

Ce mémoire de thèse concerne l’identification des facteurs de risque d’une maladie spécifique présentant une hétérogénéité spatiale au sein d’une région donnée. Plus particulièrement,nous nous sommes intéressés aux cancers des voies aéro-digestives supérieures(VADS) dans la région Nord-Pas-de-Calais (NPDC) en France. Pour cela, une étude cas témoins a d’abord été réalisée à partir de la création d’un échantillon d’individus sains qui n’étaient pas affectés par des tumeurs cancéreuses (les témoins) et d’un échantillon d’individus atteints du cancer (les cas - ou patients), recrutés dans les centres de soins dans le cadre des projets de recherche DEREDIA et NOFARIS. Concernant la méthodologie,des modèles binaires spatiaux répondant à l’objectif ont été développés à partir de travaux issus des domaines de recherche en statistique/économétrie des comportements(analyse des décisions) ainsi qu’en statistique spatiale. Un des apports méthodologiques de la thèse repose sur la combinaison de techniques provenant de ces deux champs de recherche. Dans une première partie, nous avons utilisé un modèle spatial binaire paramétrique contenant une variable spatiale latente de choix dans le cadre d’un échantillonnage des données. Ce problème est connu sous le nom de "Choice-Based Sampling" (CBS) dans les modèles discrets. Contrairement à l’échantillon aléatoire où tous les éléments de la population ont la même probabilité d’être choisi, l’échantillonnage CBS dans le modèle discret est un type d’échantillonnage dans lequel la classification de la population est faite sous forme de sous-ensembles (strates) basés sur des choix alternatifs. Dans ce contexte,l’utilisation de la procédure d’estimation par maximum de vraisemblance standard (MLE)dans le CBS pourrait mener à des estimations incohérentes (asymptotiquement biaisées).Nous avons adopté ainsi le principe du maximum de vraisemblance auprès de l’étude de scas-témoins spatiaux. Nous avons également fourni un estimateur des moments généralisés(GMM), basé sur les résidus généralisés. Dans une seconde partie, un modèle spatial binaire semi-paramétrique a été considéré. Nous présentons dans ces parties, une simulation de Monte Carlo pour étudier la performance des méthodes d’estimation au sein d’un échantillon final, que nous avons ensuite appliqué aux données du cancer VADS dans la région Nord-Pas-de-Calais. La troisième partie est consacrée à l’étude d’une fonction de risque spatiale en présence de données entachées d’erreurs. En effet, dans le cadre des données cas-témoins considérées, nous supposons que certaines données de type déclarative ne soient pas correctes. Une application de cette méthode à la cartographie du risque de développer un cancer VADS dans la région Nord-Pas-de-Calais a été étudiée. La dernière partie est consacrée à un modèle de durée spatial et son application aux données considérées. / This thesis concerns the identification of risk factors for a certain type of diseasepresenting a certain spatial heterogeneity in a given region.. More specifically, we are interested in cancers of the upper aerodigestive tract (UADT) cancers in the Nord-Pasde-Calais region (NPDC), France. For this, a case-control study was first carried out bycreating a sample of healthy individuals who are not affected by cancerous tumors (thecontrols) and a sample of individuals with cancer (Cases or patients), recruited in healthcenters as part of DEREDIA and NOFARIS research projects. From a methodologicalpoint of view, spatial binary models which meet the objective have been developed onthe basis of studies in statistical/behavioral econometrics (decision analysis) and spatialstatistics. One of the methodological contributions of the thesis on this plan is the combinationof techniques from these two fields of research.In the first part, we used a spatial binary parametric models containing spatial latentchoice variable in a context of sampling data. This problem is known as Choice-BasedSampling (CBS) in discrete choice model. Unlike the random sample where all items in the population have the same probability of being chosen, the Choice-Based Sampling indiscrete choice model is a type of sampling where the classification of the population intosubsets to be sampled is based on the choices or outcomes. In this context, the use ofstandard Maximum likelihood estimation (MLE) procedure in CBS could lead to an inconsistent(asymptotically biased) estimation. Thus, we adapt the principle of maximumlikelihood in our context of spatial case-control studies. We also provide a GMM estimatorbased on the generalized residuals.In the second part, a spatial semi-parametric binary model was considered. We present inthese parts a Monte Carlo experiment to investigate the finite sample performance of theseestimation methods, then we apply to the (UADT) cancer data in the Nord-Pas-de-Calaisregion.The third part is devoted to the study of a spatial risk function in the presence of datacontaminated by measurement errors. Indeed, in the context of the considered case-controlstudy, it is very likely that certain data transmitted by the patients is not correct. Anapplication of this method to the mapping of the risk of having UADT cancer in the Nord-Pas-de-Calais region was studied. The last part is devoted to a spatial duration modeland its application to the real data was considered.

In this study, we extend the spatial weight matrix defined by Getis and Aldstadt (2004) to a more general case. The modified spatial weight matrix performs better than the original spatial weight matrix since the modified spatial weight matrix adjusts weights of observations based on the distance from other observations. Both the simulation study and the application to the ecological process of invasion of non-native invasive plants (NNIPs) provide evidences for the better performance of the modified spatial weight matrix. We also develop procedures that can be used to quantify the invasion stages of NNIPs. The resultant map of invasion stage on county-level provides a useful and meaningful tool for policy makers; especially, it can be used to optimize allocation of management resources. The result of simultaneous autoregressive model shows that not only the biotic and abiotic factors but also human activities play an important role in the establishment and spread of multiflora rose in the Upper Midwest. It also shows the tendency of the establishment and spread of multiflora rose (Rosa Multiflora, Thunb. ex Murr.) in the Upper Midwest.

The empirical paper studies the relationship between property value and hedonic attributes. To indentify the determinant characteristics the influent the private real estate price, their degrees of significance and help with the valuation procedure, 8870 private residential property transactions with caveats lodged across country are selected from Urban Redevelopment Authority of Singapore. 40 models are tested and RMSE, R-Square, Adjusted R-Square, F-Value tests are performed to discover the overall fitness of the models. Breusch-Pagan F-Test is performed to test the existence of heteroskedasticity and VIF test to check multicolinearity. Z score is performed to check the spatial autocorrelation. Three founding are discovered. Firstly, size, age, floor level, population density level, latitude and construction status are core attributes resulting from the regression. Secondly, new district zones classified by functions are detected instead of 28 administrative districts. Thirdly, government policies and local customs (Feng Shui) are proven to be determinant variables as well. Two suggestions are given to regulate the market in the end of this article.

鑒於過去文獻可知，由於同一地區內的鄰近住宅擁有相同區位及市場特性，因而不動產價值存在高度相依性。空間相依性的產生往往是因為近鄰區域內的住宅有相似的建築結構（往往在同一個時間所興建），以及享有相同社會服務。由於建商在產品策略決策上會參考同一時間內鄰近競爭個案的產品策略，所以鄰近的新推個案會有相似的建築特徵以及相似的產品訂價。因此新推個案的訂價與鄰近的個案產生相關性，而新推個案訂價的相依性程度也會隨著時間距離遞減。 　　本文的目的在於將空間和時間的相依性最適地納入新推個案的訂價模型。採用582個台北市建商推案樣本進行實證。本研究分別以Moran’s I值和LISA值兩項指數來檢測空間自相關，並且比較傳統OLS迴歸模型、空間落遲模型，以及空間誤差模型三個模型的預測能力。此外，我們以不同的空間和時間的加權矩陣納入空間誤差模型中討論。 　　研究結果顯示，考量空間相依性之空間迴歸模型其解釋能力明顯優於一般傳統迴歸模型。而比起空間統計模型，時空迴歸模型更可以提高估計新推個案訂價的準確性。此外，研究結果亦顯示考慮時空交互影響的時空迴歸模型乃為新推個案訂價的最佳推估模式。 / It is well-known from the literature that the values of real estates are highly dependent on their locational and market characteristics of the buildings in adjacent areas. Spatial dependence mainly derives from factors that buildings at nearby properties have similar structural features (which were often developed at the same time) and often share the same social welfare. As developers in making decisions on product strategy will make reference to the strategy of nearby products of competitive cases which developed during the same time, therefore, within a certain period of time, the adjacent new construction will often have similar construction attributes as well as similar products pricing. Not only the pricing of a new construction is likely to be related to the pricing of adjacent new construction, but also the pricing of a new construction would be prone to autocorrelation decays in accordance with time distance. The aim of this paper is to analyze on how to take this temporal and spatial dependence into account in the pricing model of the new construction in the most appropriate way. We use a database of 582 asking prices of real estate developers in Taipei city. Two indices for measuring spatial autocorrelation are considered including (i) Moran’s I Index and (ii) LISA’s Index. We compared the predictive ability of three models including (i) OLS model, (ii) spatial lag model, and (iii) spatial error model. Moreover, we discussed the different temporal and spatial weight matrices in the spatial error model. According to our research results, we concluded that spatial statistical models obviously perform better than the traditional OLS model. Temporal and spatial statistical models would provide more accurate predictions on the pricing of a new construction than spatial statistical models do. The research result reveals that the best pricing model of the new construction is temporal and spatial statistical models which include temporal and spatial correlation.