In this study, a panel data set of flights made by employees at the Royal Institute of Technology (KTH) in Sweden is analyzed using generalized linear modeling approaches, with the aim to create a model with high predictive capability of the quarterly CO2 emission and the number of flights, for a year not included in the model estimation. A Zero-inflated Gamma regression model is fitted to the CO2 emission variable and a Zero-inflated Negative Binomial regression model is used for the number of flights. To build the models, cross-validation is performed with the observations from 2018 as the training set and the observations from the next year, 2019, as the test set. One at a time, the variable that best improves the prediction of the test set data (either as included in the count model or the zero-inflation model) is selected until an additional variable turns out insignificant on a 5% significance level in the estimated model. In addition to the variables in the data, three lags of the dependent variables (CO2 emission and flights) were included, as well as transformed versions of the continuous variables, and a random intercept each for the categorical variables indicating quarter and department at KTH, respectively. Neither model selected through the cross-validation process turned out to be particularly good at predicting the values for the upcoming year, but a number of variables were proven to have a statistically significant association with the respective dependent variable.

No / AIMS: To describe incidence, presentation, treatment and short term outcomes of severe neonatal hypernatraemia (SNH, sodium >/=160 mmol/l). METHODS: Prospective, population based surveillance study over 13 months using the British Paediatric Surveillance Unit. Cases were >33 weeks gestation at birth, fed breast or formula milk and <28 days of age at presentation. RESULTS: Of 62 cases of SNH reported (7, 95% CI 5.4 to 9.0 per 1 00 000 live births), 61 mothers had intended to achieve exclusive breast feeding. Infants presented at median day 6 (range 2-17) with median weight loss of 19.5% (range 8.9-30.9). 12 had jaundice and 57 weight loss as a presenting feature. 58 presented with weight loss >/=15%. 25 babies had not stooled in the 24 h prior to admission. Serum sodium fell by median 12.9 mmol/l per 24 h (range 0-30). No baby died, had seizures or coma or was treated with dialysis or a central line. At discharge, babies had regained 11% of initial birth weight after a median admission of 5 (range 2-14) days. 10 were exclusively breast fed on discharge from hospital. CONCLUSIONS: Neonatal hypernatraemia at this level, in this population, is strongly associated with weight loss. It occurs almost exclusively after attempts to initiate breast feeding, occurs uncommonly and does not appear to be associated with serious short term morbidities, beyond admission to hospital.

Dentre as possíveis propostas encontradas na literatura estatística para analisar dados oriundos de estudos com observações correlacionadas, estão os modelos condicionais e os modelos marginais. Diversas técnicas têm sido propostas para a análise de diagnóstico nesses modelos. O objetivo deste trabalho é apresentar algumas das técnicas de diagnóstico disponíveis para os dois tipos de modelos e avaliá-las por meio de estudos de simulação. As técnicas apresentadas também foram aplicadas em um conjunto de dados reais. / Conditional and marginal models are among the possibilities in statistical literature to analyze data from studies with correlated observations. Several techniques have been proposed for diagnostic analysis in these models. The objective of this work is to present some of the diagnostic techniques available for both modeling approaches and to evaluate them by simulation studies. The presented techniques were also applied in a real dataset.

OBJETIVOS: O estudo explorou diferentes métodos de séries temporais visando desenvolver um modelo para a previsão do volume diário de pacientes no Pronto Socorro do Instituto Central do Hospital das Clínicas da Faculdade de Medicina da USP. MÉTODOS: Foram explorados seis diferentes modelos para previsão do número diário de pacientes no pronto socorro de acordo com algumas variáveis relacionadas ao calendário e à temperatura média diária. Para a construção dos modelos, utilizou-se a contagem diária de pacientes atendidos no pronto socorro entre 1° de janeiro de 2008 a 31 de dezembro de 2010. Os primeiros 33 meses do banco de dados foram utilizados para o desenvolvimento e ajuste dos modelos, e os últimos três meses foram utilizados para comparação dos resultados obtidos em termos da acurácia de previsão. A acurácia foi medida a partir do erro médio percentual absoluto. Os modelos foram desenvolvidos utilizando-se três diferentes métodos: modelos lineares generalizados, equações de estimação generalizadas e modelos sazonais autorregressivos integrados de média móvel (SARIMA). Para cada método, foram testados modelos que incluíram termos para controlar o efeito da temperatura média diária e modelos que não incluíram esse controle. RESULTADOS: Foram atendidos, em média, 389 pacientes diariamente no pronto socorro, número que variou entre 166 e 613. Observou-se uma sazonalidade semanal marcante na distribuição do volume de pacientes ao longo do tempo, com maior número de pacientes às segundas feiras e tendência linear decrescente ao longo da semana. Não foi observada variação significante no volume de pacientes de acordo com os meses do ano. Os modelos lineares generalizados e equações de estimação generalizada resultaram em melhor acurácia de previsão que os modelos SARIMA. No primeiro horizonte de previsão (outubro), por exemplo, os erros médios percentuais absolutos dos modelos lineares generalizados e de equação de estimação generalizada foram ambos 11,5% e 10,8% (modelos que incluíram e que não incluíram termo para controlar o efeito da temperatura, respectivamente), enquanto os erros médios percentuais absolutos para os modelos SARIMA foram 12,8% e 11,7% (modelos que incluíram e que não incluíram termo para controlar o efeito da temperatura, respectivamente). Para todos os modelos, incluir termos para controlar o efeito da temperatura média diária não resultou em melhor acurácia de previsão. A previsão a curto prazo (7 dias) em geral resultou em maior acurácia do que a previsão a longo prazo (30 dias). CONCLUSÕES: Este estudo indica que métodos de séries temporais podem ser aplicados na rotina do serviço de pronto socorro para a previsão do provável volume diário de pacientes no serviço. A previsão realizada para o curto prazo tem boa acurácia e pode ser incorporada à rotina do serviço, de modo a subsidiar seu planejamento e colaborar com a adequação de recursos materiais e humanos. Os modelos de previsão baseados unicamente em variáveis relacionadas ao calendário foram capazes de prever a variação no volume diário de pacientes, e os métodos aqui aplicados podem ser automatizados para gerar informações com antecedência suficiente para decisões de planejamento do serviço de pronto socorro / OBJECTIVES: This study aims to develop different models to forecast the daily number of patients seeking emergency department (ED) care in a general hospital according to calendar variables and ambient temperature readings and to compare the models in terms of forecasting accuracy. METHODS: We developed and tested six different models of ED patient visits using total daily counts of patient visits to the Instituto Central do Hospital das Clínicas Emergency Department from January 1, 2008 to December 31, 2010. We used the first 33 months of the dataset to develop the ED patient visits forecasting models (the training set), leaving the last 3 months to measure each model\'s forecasting accuracy by the mean absolute percentage error. Forecasting models were developed using 3 different time series analysis methods: generalized linear models, generalized estimating equations and seasonal autoregressive integrated moving average (SARIMA). For each method, we explored models with and without the effect of mean daily temperature as a predictive variable. RESULTS: Daily mean number of ED visits was 389, ranging from 166 to 613. Data showed a weekly seasonal distribution, with highest patient volumes on Mondays and lowest patient volumes on weekends. There was little variation in daily visits by month. Generalized linear models and generalized estimating equation models showed better forecasting accuracy than SARIMA models. For instance, the mean absolute percentage errors from generalized linear models and generalized estimating equations models at the first month of forecasting (October, 2012), were 11.5% and 10.8% (models with and without control for the temperature effect, respectively), while the mean absolute percentage errors from SARIMA models were 12.8% and 11.7% (models with and without control for the temperature effect, respectively). For all models, controlling for the effect of temperature resulted in worse or similar forecasting ability than models with calendar variables alone, and forecasting accuracy was better for the short term horizon (7 days in advance) than for the longer term (30 days in advance). CONCLUSIONS: Our study indicates that time series models can be developed to provide forecasts of daily ED patient visits, and forecasting ability was dependent on the type of model employed and the length of the time-horizon being predicted. In our setting, generalized linear models and generalized estimating equation models showed better accuracy, and including information about ambient temperature in the models did not improve forecasting accuracy. Forecasting models based on calendar variables alone did in general detect patterns of daily variability in ED volume, and thus could be used for developing an automated system for better planning of personnel resources

Diversos estudos têm sido feitos para verificar quais as relações entre a lagarta do cartucho-do-milho, Spodoptera frugiperda, bem como o curuquerê-do-algodoeiro, Alabama argillacea, e potenciais agentes de controle biológico, visto que são importantes pragas de diversas culturas. Pesquisas com parasitoides e predadores de diferentes ordens têm sido desenvolvidas, com esse propósito. A tesourinha Euborellia annulipes apresenta grande potencial, assim como o pentatomídeo Podisus nigrispinus, por serem inimigos naturais desses insetos-praga. O parasitoide Campoletis flavicincta é, também, considerado um potencial agente controlador de S. frugiperda, no campo. Quatro experimentos foram instalados e conduzidos para estudar as relações ecológicas entre as espécies citadas: i) competição entre os predadores por larvas de S. frugiperda; ii) competição entre os predadores por larvas de A. argillacea; iii) competição intraespecífica entre as ninfas dos predadores; iv) teste de preferência por larvas de S. frugiperda previamente parasitadas, ou não, por C. flavicincta. Modelos estatísticos de diferentes tipos foram ajustados aos dados obtidos. Foram utilizados modelos lineares generalizados para dados discretos univariados (contagens e proporções), modelos de análise de sobrevivência para dados de tempo até ocorrência de um determinado evento, modelos aditivos generalizados e modelos multinomiais ordinais para dados de taxas de mortalidade observadas ao longo do tempo. Para verificar a qualidade do ajuste de modelos lineares generalizados, funções que produzem gráficos meio-normais com envelope de simulação foram implementadas no software estatístico gratuito R. Observou-se que as fêmeas de E. annulipes são mais agressivas e a espécie tem vantagem competitiva sobre P. nigrispinus, podendo atuar como predadora intraguilda. Além disso, os predadores têm a capacidade de discriminar entre larvas de S. frugiperda parasitadas ou não por C. flavicincta, sendo que há evidências de que os machos evitam consumir larvas parasitadas. Já as fêmeas não têm preferência e podem consumir o parasitoide, indiretamente, ao predar uma larva parasitada, sendo que isso pode ser um fator estabilizador do sistema. Adicionalmente, a coexistência entre a praga, os predadores e o parasitoide é mais provável quando as taxas de ataque dos predadores são mais altas e a do parasitoide é mais baixa. / Several studies have verified the relations among the fall armyworm, Spodoptera frugiperda, as well as the cotton leafworm, Alabama argillacea, and potential biological control agents, as they are important pests of several crops. Research with parasitoids and predators of different orders has been developed with this purpose. The ring-legged earwig (Euborellia annulipes) shows great potential, as well as the Neotrpical stink bug (Podisus nigrispinus), as they are natural enemies of these pests. The parasitoid wasp Campoletis flavicincta is also considered a biological control agent of S. frugiperda. Four experiments were set up and conducted to characterize the ecological relations among the described species: i) competition between the predators for S. frugiperda larvae; ii) competition between the predators for A. argillacea larvae; iii) intraspecific competition among the predators\' nymphs; iv) choice tests between S. frugiperda larvae that were previously parasitised by C. flavicincta and larvae that were not. Statistical models of different types were fit to the data. Generalized linear models were used to analyse univariate discrete data (counts and proportions), survival analysis models were used to analyse time-until-event data, generalized additive models and ordered multinomial logistic models were used to analyse mortality rates through time. To assess goodness-of-fit, functions that produce half-normal plots with a simulation envelope were implemented for the open-source software R. It was observed that females of E. annulipes are less aggressive and this species has a competitive advantage over P. nigrispinus, and may act as an intraguild predator. Besides that, the predators are able to distinguish parasitised from non-parasitised S. frugiperda larvae, and there are evidence that show that males tend to avoid feeding on larvae parasitised by C. flavicincta. Females, however, have no prefference and may indirectly feed on the parasitoid when predating a parasitised larva, and this might be a stabilizing factor in this system. In addition, coexistence among the pest, predators and parasitoid is most likely when predation rates are high and parasitism rates are low.

In der Arbeit wird eine Methodik einer zusammenhängenden Analyse und modellhaften Beschreibung der Verkehrssicherheit in städtischen Hauptstraßennetzen am Beispiel der Stadt Dresden entwickelt. Die dabei gewonnenen Modelle dienen der Abschätzung von Erwartungswerten von Unfallhäufigkeiten mit und ohne Personenschaden unter Berücksichtigung der Verkehrsbeteiligungsart. Die Grundlage bilden multivariate Regressionsmodelle auf Basis verallgemeinerter linearer Modelle (GLM). Die Verwendung verallgemeinerter Regressionsmodelle erlaubt eine Berücksichtigung von Verteilungen, die besser geeignet sind, den Unfallentstehungsprozess wiederzugeben, als die häufig verwendete Normalverteilung. Im konkreten Fall werden hierzu die Poisson-Verteilung sowie die negative Binomialverteilung verwendet. Um Effekte im Hauptverkehrsstraßennetz möglichst trennscharf abbilden zu können, werden vier grundsätzliche Netzelemente differenziert und das Netz entsprechend zerlegt. Unterschieden werden neben Streckenabschnitten und Hauptverkehrsknotenpunkten auch Annäherungsbereiche und Anschlussknotenpunkte. Die Kollektive der Knotenpunkte werden ferner in signalisierte und nicht-signalisierte unterteilt. Es werden zunächst Modelle unterschiedlicher Unfallkollektive getrennt für alle Kollektive der vier Netzelemente berechnet. Anschließend werden verschiedene Vorgehensweisen für eine Zusammenfassung zu Netzmodellen entwickelt. Neben der Verwendung verkehrstechnischer und infrastruktureller Größen als erklärende Variable werden in der Arbeit auch Kenngrößen zur Beschreibung der Umfeldnutzung ermittelt und im Rahmen der Regression einbezogen. Die Quantifizierung der Umfeldnutzung erfolgt mit Hilfe von Korrelations-, Kontingenz- und von Hauptkomponentenanalysen (PCA). Im Ergebnis werden Modelle präsentiert, die eine multivariate Quantifizierung erwarteter Unfallhäufigkeiten in Hauptverkehrsstraßennetzen erlauben. Die vorgestellte Methodik bildet eine mögliche Grundlage für eine differenzierte Sicherheitsbewertung verkehrsplanerischer Variantenabschätzungen. / A methodology is developed in order to predict the number of accidents within an urban main road network. The analysis was carried out by surveying the road network of Dresden. The resulting models allow the calculation of individual expectancy values for accidents with and without injury involving different traffic modes. The statistical modelling process is based on generalized linear models (GLM). These were chosen due to their ability to take into account certain non-normal distributions. In the specific case of accident counts, both the Poisson distribution and the negative binomial distribution are more suitable for reproducing the origination process than the normal distribution. Thus they were chosen as underlying distributions for the subsequent regressions. In order to differentiate overlaying influences, the main road network is separated into four basic elements: major intersections, road sections, minor intersections and approaches. Furthermore the major and minor intersections are additionally subdivided into signalised and non-signalised intersections. Separate models are calculated for different accident collectives for the various types of elements. Afterwards several methodologies for calculating aggregated network models are developed and analysed. Apart from traffic-related and infrastructural attributes, environmental parameters are derived taking into account the adjacent building structure as well as the surrounding land-use, and incorporated as explanatory variables within the regression. The environmental variables are derived from statistical analyses including correlation matrices, contingency tables and principal components analyses (PCA). As a result, a set of models is introduced which allows a multivariate calculation of expected accident counts for urban main road networks. The methodology developed can serve as a basis for a differentiated safety assessment of varying scenarios within a traffic planning process.

Ο Μαυροπετρίτης (Falco eleonorae Géné, 1839) είναι ένα μεσαίου μεγέθους μεταναστευτικό γεράκι το οποίο επισκέπτεται τη χώρα μας τους καλοκαιρινούς μήνες για να αναπαραχθεί. Οι αναπαραγωγικές αποικίες εντοπίζονται σε νησιά και ακτές της Μεσογείου, της Μακαρονησίας και της βορειοδυτικής Αφρικής, ενώ τους υπόλοιπους μήνες το είδος απαντά στη Μαδαγασκάρη και γειτονικές περιοχές. Στο παρελθόν έχει πραγματοποιηθεί ένας μεγάλος αριθμός μελετών σχετικά με την περίοδο αναπαραγωγής, ενώ για τον υπόλοιπο κύκλο ζωής του είδους οι γνώσεις μας μέχρι τις αρχές του 21ου αιώνα ήταν περιορισμένες. Στην παρούσα διατριβή μελετώνται τα βασικότερα στάδια του κύκλου ζωής του Μαυροπετρίτη, επικεντρώνοντας στην περίοδο αναπαραγωγής, μετανάστευσης και διαχείμασης ατόμων που αναπαράγονται στην Ελλάδα. Συγκεκριμένα, διερευνάται (α) η αναπαραγωγική επιτυχία του είδους και η επίδραση περιβαλλοντικών παραγόντων στην έκβαση της αναπαραγωγικής επιτυχίας, (β) το πρότυπο κατανομής των αναπαραγωγικών αποικιών στο Αιγαίο και Ιόνιο πέλαγος, (γ) τα κριτήρια που χρησιμοποιούν τα αναπαραγωγικά ζευγάρια κατά την επιλογή θέσης φωλιάσματος σε ακατοίκητες νησίδες του Αιγαίου πελάγους, (δ) η περίοδος μετανάστευσης τεσσάρων γερακιών από μια αποικία του Κεντρικού Αιγαίου και οι παράγοντες που επηρεάζουν το πρότυπο μετανάστευσής τους και, (ε) η περίοδος διαχείμασης και το πρότυπο κατανομής των τεσσάρων γερακιών στην περιοχή διαχείμασης. Η εκπόνηση της παρούσας διατριβής βασίσθηκε σε μεγάλο βαθμό στα πρωτογενή δεδομένα που συλλέχθηκαν στο πλαίσιο του προγράμματος «LIFE - ΦΥΣΗ 2003 Δράσεις για την προστασία του Μαυροπετρίτη (Falco eleonorae) στην Ελλάδα (LIFE 03NAT/GR/000091)», με κύριο ανάδοχο την Ελληνική Ορνιθολογική Εταιρεία (Ε.Ο.Ε.) σε συνεργασία με το Μουσείο Φυσικής Ιστορίας Κρήτης (ΜΦΙΚ), το Υπουργείο Αγροτικής Ανάπτυξης και Τροφίμων και τη Βρετανική Ορνιθολογική Εταιρεία (RSPB) και με χρηματοδότηση της Ευρωπαϊκής Επιτροπής και συγχρηματοδότηση του ιδρύματος «Α. Γ.Λεβέντης». Το πρόγραμμα παρακολούθησης της μετανάστευσης των τεσσάρων Μαυροπετριτών πραγματοποιήθηκε στο πλαίσιο του προγράμματος «Μελέτη των πτηνών του ελληνικού θαλάσσιου χώρου» με ανάδοχο την Ελληνική Ορνιθολογική Εταιρεία και με χρηματοδότηση του ιδρύματος «Α. Γ. Λεβέντης». / Eleonora's falcon (Falco eleonorae Géné, 1839) is a medium-sized migratory raptor that visits the Greek islands during the breeding season. Its breeding colonies are distributed on islands and coasts of the Mediterranean Sea, Macaronesia and northwest Africa, while during the rest of the year the species is found in Madagascar and surrounding islets. In previous years most studies focused on the breeding period, while up to the 21st century our knowledge with regards to the rest of its life cycle remained relatively poor. In this thesis i study the main stages of Eleonora's falcon life cycle, focusing on the breeding, migratory and wintering period of individuals that breed in Greece. In particular, i investigate (a) the breeding success of the species and the effect of environmental factors on the outcome of the breeding effort, (b) the distribution pattern of the breeding colonies in the Aegean and Ionean Sea, (c) the criteria the breeding pairs use at the time of nesting site selection in uninhabited islets of the Aegean Sea, (d) the migratory period of four falcons originating from a breeding colony of the Central Aegean Sea and the factors that shape the irmigratory pattern and, (e) the wintering period and the distribution pattern of the four falcons in their wintering grounds. This thesis was based to a great extent on the data collected in the frame of the project "LIFE-Nature 2003 Conservation Measures for Falco eleonorae in Greece (LIFE 03NAT/GR/000091)" undertaken by the Hellenic Ornithological Society (H.O.S.) in collaboration with the Natural History Museum of Crete (NHMC), the Ministry of Rural Development and Food and the Royal Society for the Protection of Birds (RSPB), which was funded by the European Commission and cofunded by the A.G. Leventis Foundation. The migration tracking project was funded by the A.G. Leventis Foundation through project "Survey and Conservation of Seabirds in Greece".

The term aggregation refers to overdispersion and both are used interchangeably in this thesis. In addressing the problem of prevalence of infectious parasite species faced by most rural livestock farmers, we model the distribution of faecal egg counts of 15 parasite species (13 internal parasites and 2 ticks) common in sheep and goats. Aggregation and excess zeroes is addressed through the use of generalised linear models. The abundance of each species was modelled using six different distributions: the Poisson, negative binomial (NB), zero-inflated Poisson (ZIP), zero-inflated negative binomial (ZINB), zero-altered Poisson (ZAP) and zero-altered negative binomial (ZANB) and their fit was later compared. Excess zero models (ZIP, ZINB, ZAP and ZANB) were found to be a better fit compared to standard count models (Poisson and negative binomial) in all 15 cases. We further investigated how distributional assumption a↵ects aggregation and zero inflation. Aggregation and zero inflation (measured by the dispersion parameter k and the zero inflation probability) were found to vary greatly with distributional assumption; this in turn changed the fixed-effects structure. Serial autocorrelation between adjacent observations was later taken into account by fitting observation driven time series models to the data. Simultaneously taking into account autocorrelation, overdispersion and zero inflation proved to be successful as zero inflated autoregressive models performed better than zero inflated models in most cases. Apart from contribution to the knowledge of science, predictability of parasite burden will help farmers with effective disease management interventions. Researchers confronted with the task of analysing count data with excess zeroes can use the findings of this illustrative study as a guideline irrespective of their research discipline. Statistical methods from model selection, quantifying of zero inflation through to accounting for serial autocorrelation are described and illustrated. / Statistics / M.Sc. (Statistics)

This thesis is focused on environmental preferences of wood-decaying fungi and their relationships with forest structure and development. Relationships of fungi to properties of wood and forest stands were studied on the basis of field observations in Central-European mountain spruce forests. Plot-based approach was used to reveal a general pattern in the diversity of fungi within a single forest stand and between different stands. The analysis of stand structure provided a background for plot-based approach. Substrate-based approach was used to study single species preferences and their communities. In addition, the influence of wood properties (including fungi and their rots) on the regeneration of spruce on logs was studied.

Made available in DSpace on 2016-06-02T20:06:02Z (GMT). No. of bitstreams: 1 2447.pdf: 2730026 bytes, checksum: 0c7b575bbfeb3fed2fc6c929b9785516 (MD5) Previous issue date: 2009-03-05 / Financiadora de Estudos e Projetos / The objective of this work is to present techniques of regression analysis for longitudinal data when the response variable is binary. Initially, there is a review of generalized linear models, marginal models, transition models, mixed models, and logistic regression methods of estimation, which will be necessary for the development of work. In addition to the methods of estimation, some structures of correlation will be studied in an attempt to capture the intra-individual serial dependence over time. These methods were applied in two situations, one where the response variable is continuous and normal distribution, and another when the response variable has the Bernoulli distribution. It was also sought to explore and present techniques for selection of models and diagnostics for the two cases. Finally, an application of the above methodology will be presented using a set of real data. / O objetivo deste trabalho é apresentar técnicas de análise de regressão para dados longitudinais quando a variável resposta é binária. Inicialmente, é feita uma revisão sobre modelos lineares generalizados, modelos marginais, modelos de transição, modelos mistos, regressão logística e métodos de estimação, pois serão necessários para o desenvolvimento do trabalho. Além dos métodos de estimação, algumas estruturas de correlação serão estudadas, na tentativa de captar a dependência serial intra-indivíduo ao longo do tempo. Estes métodos foram aplicados em duas situações; uma quando a variável resposta é contínua, e se assume ter distribuição normal, e a outra quando a variável resposta assume ter distribuição de Bernoulli. Também se procurou pesquisar e apresentar técnicas de seleção de modelos e de diagnósticos para os dois casos. Ao final, uma aplicação com a metodologia pesquisada será apresentada utilizando um conjunto de dados reais.