Background and Objectives: I investigated the use of sensitivity analyses in assessing statistical results or analytical approaches in three different statistical issues: (1) accounting for within-subject correlations in analyzing discrete choice data, (2) handling both-armed zero-event studies in meta-analyses for rare event outcomes, and (3) incorporating external information using Bayesian approach to estimate rare-event rates. Methods: Project 1: I empirically compared ten statistical models in analyzing correlated data from a discrete choice survey to elicit patient preference for colorectal cancer screening. Logistic and probit models with random-effects, generalized estimating equations or robust standard errors were applied to binary, multinomial or bivariate outcomes. Project 2: I investigated the impacts of including or excluding both-armed zero-event studies on pooled odds ratios for classical meta-analyses using simulated data. Five commonly used pooling methods: Peto, Mantel-Haenszel fixed/random effects and inverse variance fixed/random effects, were compared in terms of bias and precision. Project 3: I explored the use of Bayesian approach to incorporate external information through priors to verify, enhance or modify the study evidence. Three study scenarios were derived from previous studies to estimate inhibitor rates for hemophilia A patients treated with rAHF-PFM: 1) a single cohort of previously treated patients, 2) individual patient data meta-analysis, and 3) an previously unexplored patient population with limited data. Results and Conclusion: Project 1: When within-subject correlations were substantial, the results from different statistical models were inconsistent. Project 2: Including both-armed zero-event studies in meta-analyses increased biases for pooled odd ratios when true treatment effects existed. Project 3: Through priors, Bayesian approaches effectively incorporated different types of information to strengthen or broaden research evidence. Through this thesis I demonstrated that when analyzing complicated health research data, it was important to use sensitivity analyses to assess the robustness of analysis results or proper choice of statistical models. / Dissertation / Doctor of Philosophy (PhD)

Proposer une offre en logements adaptée aux différents besoins et préférences des ménages représente un enjeu important pour les acteurs publics de l’aménagement. Ces besoins et préférences dépendent des caractéristiques des ménages et des changements qu’ils peuvent connaître dans leur cycle de vie (mise en couple, naissance, séparation…). Les facteurs participant aux choix résidentiels sont nombreux (attributs du logement, caractéristiques de l’environnement résidentiel) et interviennent différemment selon les types de ménages. Les dynamiques résidentielles impliquent ainsi une grande variété d’éléments, en interaction les uns avec les autres, et les relations de cause à effet sont difficiles à identifier. Par conséquent, il n’est pas possible de prévoir le comportement résidentiel des ménages pas plus que leurs évolutions possibles sans outil adapté.Pour étudier les dynamiques résidentielles intra-urbaines, nous utilisons dans cette thèse un modèle de simulation des mobilités résidentielles (Mobisim-MR) intégré dans une plateforme de simulation LUTI individu-centrée : Mobisim. Mobisim-MR permet de déterminer, pour chaque année de simulation, les ménages qui déménagent et leur nouvelle localisation résidentielle. En amont de Mobisim-MR, un modèle de simulation des évolutions démographiques (Mobisim-Démo) a été créé au sein de la plateforme Mobisim. Il permet de reproduire de manière dynamique et individu-centrée l’évolution des ménages dans leur cycle de vie. Une partie de la thèse est dédiée au paramétrage ces deux modèles, étape préalable nécessaire à la simulation de scénarios. Un autre volet de la thèse concerne l’exploration du comportement du modèle Mobisim-MR pour évaluer la stabilité des résultats de simulation et leur cohérence (analyse de sensibilité). L’utilisation de modèles individu-centrés est relativement récente en géographie, d’où l’absence de protocole standard pour l’exploration de tels modèles. Un protocole spécifique a été conçu pour explorer le comportement de Mobisim-MR. Ce protocole tient compte de la nature des paramètres du modèle, des contraintes techniques de simulation et de l’objectif pour lequel le modèle a été conçu.Le dernier volet de la thèse consiste en des analyses thématiques visant à étudier l’impact de deux scénarios de politiques de construction de logements sur l’agglomération du Grand Besançon. Ces analyses montrent la capacité de Mobisim-MR à répondre à des questions concrètes d’aménagement et à apporter des éléments de discussion aux acteurs publics en charge des politiques de logement. / To ensure that housing supply is suitable to households’ needs and preferences represents a major planning concern. These needs and preferences depend on the households’ characteristics and on their lifecycle changes (union, birth, divorce…). Residential choice factors are numerous (housing and residential environment characteristics) and their role is often different according to the types of households. Residential dynamics involve a great variety of elements, in interaction with each other, and the causal relationships are difficult to identify. Thus, it is not possible to predict the households’ residential behaviour, nor their possible evolutions, without a suitable tool. To study intra-urban residential dynamics, we use a residential mobility simulation model (Mobisim-MR), integrated in an agent-based LUTI simulation platform: Mobisim. For each simulated year, Mobisim-MR allows for determination of households which move and their new residential location. Prior to Mobisim-MR, we created a demographic microsimulation model (Mobisim-Démo) within the Mobisim platform. It allows reproducing households lifecycle evolutions in a dynamic and agent-based way. A part of the thesis is dedicated to the calibration of both models, a required stage preliminary to scenarios simulation. Another part of the thesis concerns the exploration of Mobisim-MR model behaviour, in order to assess the simulation results’ stability and their consistency (sensitivity analysis). Agent-based models use is quite recent in geography, explaining the lack of standard protocol to explore such models. A specific protocol has been designed to explore the behaviour of Mobisim-MR. This protocol takes into consideration the parameters characteristics, simulation technical constraints, and the initial design for which the model has been built.The last part of the thesis consists of thematic analyses aimed at studying the impact of two housing construction planning scenarios in the urban region of Besançon (named le Grand Besançon). These analyses highlight the ability of Mobisim-MR to answer concrete planning questions and to initiate discussion among urban planners.

Le concept de réseau d'interactions est central en Écologie et différents modèles et méthodes ont été utilisés. Ce travail de thèse met en relation deux approches développées par des courants séparés : l'étude des matrices de communauté et les analyses entrée-sortie. Il confronte leurs hypothèses aux propriétés fondamentales des écosystèmes estuariens. Il précise explicitement les liens entre les différentes matrices, ainsi que la signification des mixed trophic impacts. Les matrices de la storage et de la throughflowanalyses sont reliées à des jacobiennes de modèles à compartiments dont les flux sont contrôlés par les compartiments donneurs ou receveurs. Contrairement à ce qui est le plus souvent présenté dans la littérature, l'analyse des mixed trophic impacts est ici interprétée en terme de présence-absence du compartiment considéré. Avec les données disponibles sur les réseaux trophiques de cinq estuaires européens, des matrices qualitatives et quantitatives sont construites afin de réaliser des analyses de sensibilité. Ces premières explorations montrent des niveaux d'incertitudes très élevés, et ceci même pour le signe des prédictions. Par ailleurs, ce travail de thèse approfondit les possibilités d'étude de dynamiques transitoires à partir de la matrice de communauté. Il souligne des éléments importants qu'il est nécessaire de prendre en compte lorsque ces approches sont choisies. / Networks are a key concept in ecology and a number of models and methods have been used. This PhD dissertation links two approaches, the community matrix and input-output analyses, which have been developed by separate streams of theory. It compares their assumptions with important features of estuarine systems. It explicitly analyses the links between the matrices and the significance of the mixed trophic impacts analysis. Matrices of storage and throughflow analyses are linked to Jacobian matrices of donor or recipient controlled compartment models. Unlike most of what can be seen in the litterature, here the mixed trophic impacts are interpreted as the effects of a compartment being present or absent. Using available data in the case of five European estuaries, qualitative and quantitative matrices are built in the aim of performing sensitivity analyses. First explorations reveal high levels of uncertainties, even in the sign of the predictions. Furthermore, this work examines in more details the possibilities to explore transient dynamics from the community matrix. This PhD dissertation emphasises important features which are necessary to consider when choosing such approaches.

Clostridium thermocellum is a thermophilic anaerobe that is capable of producing ethanol directly from lignocellulosic compounds, however this organism suffers from low ethanol tolerance and low ethanol yields. In vivo mathematical modeling studies based on steady state traditional metabolic flux analysis, metabolic control analysis, transient and steady states’ flux spectrum analysis (FSA) were conducted on C. thermocellum’s central metabolism. The models were developed in Matrix Laboratory software ( MATLAB® (The Language of Technical Computing), R2008b, Version 7.7.0.471)) based on known stoichiometry from C. thermocellum pathway and known physical constraints. Growth on cellobiose from Metabolic flux analysis (MFA) and Metabolic control analysis (MCA) of wild type (WT) and ethanol adapted (EA) cells showed that, at lower than optimum exogenous ethanol levels, ethanol to acetate (E/A) ratios increased by approximately 29% in WT cells and 7% in EA cells. Sensitivity analyses of the MFA and MCA models indicated that the effects of variability in experimental data on model predictions were minimal (within ±5% differences in predictions if the experimental data varied up to ±20%). Steady state FSA model predictions showed that, an optimum hydrogen flux of ~5mM/hr in the presence of pressure equal to or above 7MPa inhibits ferrodoxin hydrogenase which causes NAD re-oxidation in the system to increase ethanol yields to about 3.5 mol ethanol/mol cellobiose.

No description available.

abstract: In the field of infectious disease epidemiology, the assessment of model robustness outcomes plays a significant role in the identification, reformulation, and evaluation of preparedness strategies aimed at limiting the impact of catastrophic events (pandemics or the deliberate release of biological agents) or used in the management of disease prevention strategies, or employed in the identification and evaluation of control or mitigation measures. The research work in this dissertation focuses on: The comparison and assessment of the role of exponentially distributed waiting times versus the use of generalized non-exponential parametric distributed waiting times of infectious periods on the quantitative and qualitative outcomes generated by Susceptible-Infectious-Removed (SIR) models. Specifically, Gamma distributed infectious periods are considered in the three research projects developed following the applications found in (Bailey 1964, Anderson 1980, Wearing 2005, Feng 2007, Feng 2007, Yan 2008, lloyd 2009, Vergu 2010). i) The first project focuses on the influence of input model parameters, such as the transmission rate, mean and variance of Gamma distributed infectious periods, on disease prevalence, the peak epidemic size and its timing, final epidemic size, epidemic duration and basic reproduction number. Global uncertainty and sensitivity analyses are carried out using a deterministic Susceptible-Infectious-Recovered (SIR) model. The quantitative effect and qualitative relation between input model parameters and outcome variables are established using Latin Hypercube Sampling (LHS) and Partial rank correlation coefficient (PRCC) and Spearman rank correlation coefficient (RCC) sensitivity indices. We learnt that: For relatively low (R0 close to one) to high (mean of R0 equals 15) transmissibility, the variance of the Gamma distribution for the infectious period, input parameter of the deterministic age-of-infection SIR model, is key (statistically significant) on the predictability of the epidemiological variables such as the epidemic duration and the peak size and timing of the prevalence of infectious individuals and therefore, for the predictability these variables, it is preferable to utilize a nonlinear system of Volterra integral equations, rather than a nonlinear system of ordinary differential equations. The predictability of epidemiological variables such as the final epidemic size and the basic reproduction number are unaffected by (or independent of) the variance of the Gamma distribution for the infectious period and therefore for the choice on which type of nonlinear system for the description of the SIR model (VIE's or ODE's) is irrelevant. Although, for practical proposes, with the aim of lowering the complexity and number operations in the numerical methods, a nonlinear system of ordinary differential equations is preferred. The main contribution lies in the development of a model based decision-tool that helps determine when SIR models given in terms of Volterra integral equations are equivalent or better suited than SIR models that only consider exponentially distributed infectious periods. ii) The second project addresses the question of whether or not there is sufficient evidence to conclude that two empirical distributions for a single epidemiological outcome, one generated using a stochastic SIR model under exponentially distributed infectious periods and the other under the non-exponentially distributed infectious period, are statistically dissimilar. The stochastic formulations are modeled via a continuous time Markov chain model. The statistical hypothesis test is conducted using the non-parametric Kolmogorov-Smirnov test. We found evidence that shows that for low to moderate transmissibility, all empirical distribution pairs (generated from exponential and non-exponential distributions) for each of the epidemiological quantities considered are statistically dissimilar. The research in this project helps determine whether the weakening exponential distribution assumption must be considered in the estimation of probability of events defined from the empirical distribution of specific random variables. iii) The third project involves the assessment of the effect of exponentially distributed infectious periods on estimates of input parameter and the associated outcome variable predictions. Quantities unaffected by the use of exponentially distributed infectious period within low transmissibility scenarios include, the prevalence peak time, final epidemic size, epidemic duration and basic reproduction number and for high transmissibility scenarios only the prevalence peak time and final epidemic size. An application designed to determine from incidence data whether there is sufficient statistical evidence to conclude that the infectious period distribution should not be modeled by an exponential distribution is developed. A method for estimating explicitly specified non-exponential parametric probability density functions for the infectious period from epidemiological data is developed. The methodologies presented in this dissertation may be applicable to models where waiting times are used to model transitions between stages, a process that is common in the study of life-history dynamics of many ecological systems. / Dissertation/Thesis / Ph.D. Applied Mathematics for the Life and Social Sciences 2014

Submitted by isabela.moljf@hotmail.com (isabela.moljf@hotmail.com) on 2017-03-02T12:45:01Z No. of bitstreams: 1 arleilucasdesouzarosa.pdf: 827143 bytes, checksum: 9b9df97bf2f2804c9287dcee04958895 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-03-06T19:38:30Z (GMT) No. of bitstreams: 1 arleilucasdesouzarosa.pdf: 827143 bytes, checksum: 9b9df97bf2f2804c9287dcee04958895 (MD5) / Made available in DSpace on 2017-03-06T19:38:30Z (GMT). No. of bitstreams: 1 arleilucasdesouzarosa.pdf: 827143 bytes, checksum: 9b9df97bf2f2804c9287dcee04958895 (MD5) Previous issue date: 2009-05-27 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Durante décadas, problemas relacionados à estabilidade de tensão ou, segurança de tensão, têm despertada a atenção crescente de pesquisadores do setor elétrico. O crescimento da economia e o aumento do consumo de energia elétrica, aliados à falta de investimentos nos setores de geração e transmissão, têm levado o sistema elétrico de potência a operar muito próximo de seus limites. Desta forma, a habilidade do sistema em manter um nível operacional satisfatório, após a ocorrência de alguma contingência, pode ficar comprometida. Este trabalho avalia a segurança de tensão através de um método simples, eficiente e rápido computacionalmente, além de propor ações de reforço para melhorar a margem de carregamento do sistema elétrico. A análise da rede é função dos caminhos de transmissão de potência ativa e um modelo matemático muito simples é utilizado. Quanto às ações de reforço, a análise de sensibilidade fornece as barras mais adequadas ao redespacho de potência ativa e reativa. Um processo iterativo é estabelecido com o intuito de recomendar ações de controle e fornecer condições de segurança de tensão menos severas. Os resultados apresentados para um sistema real de 34 barras demonstram a eficácia da metodologia desenvolvida. / Voltage stability or voltage security analysis has motivated an expressive attention of power systems researchers. The economy growth and the increased electric energy consumption, allied to the lack of investments in generation and transmission areas, have led electrical power systems to operate very close to their limits. Therefore, they cannot be able to operate in a secure way after some contingencies. In this work, a simple and fast method evaluates voltage security and proposes conditions to improve the loading margin. System analysis is carried out in terms of active power transmission path. Sensitivity analysis provides the most adequate buses to active and reactive power redispatch. A sequential iterative methodology to reinforce system conditions is presented. The results for a real 34-bus system demonstrate the efficiency of proposed methodology.

Using a bioeconomic model and stochastic simulation to assess the economic viability of small-scale, recirculating shrimp farms in the Midwestern U.S. A series of stress tests were implemented on key input variables including survival rate, selling price, electricity usage, discount rate and the cost of added salt. The key output variable is the Net Present Value of the operation. <div><br></div><div><br></div>