Adaptive Markov chain Monte Carlo and Bayesian filtering for state space models

This thesis is concerned with the state and parameter estimation in state space models. The estimation of states and parameters is an important task when mathematical modeling is applied to many different application areas such as the global positioning systems, target tracking, navigation, brain imaging, spread of infectious diseases, biological processes, telecommunications, audio signal processing, stochastic optimal control, machine learning, and physical systems. In Bayesian settings, the estimation of states or parameters amounts to computation of the posterior probability density function. Except for a very restricted number of models, it is impossible to compute this density function in a closed form. Hence, we need approximation methods. A state estimation problem involves estimating the states (latent variables) that are not directly observed in the output of the system. In this thesis, we use the Kalman filter, extended Kalman filter, Gauss–Hermite filters, and particle filters to estimate the states based on available measurements. Among these filters, particle filters are numerical methods for approximating the filtering distributions of non-linear non-Gaussian state space models via Monte Carlo. The performance of a particle filter heavily depends on the chosen importance distribution. For instance, inappropriate choice of the importance distribution can lead to the failure of convergence of the particle filter algorithm. In this thesis, we analyze the theoretical Lᵖ particle filter convergence with general importance distributions, where p ≥2 is an integer. A parameter estimation problem is considered with inferring the model parameters from measurements. For high-dimensional complex models, estimation of parameters can be done by Markov chain Monte Carlo (MCMC) methods. In its operation, the MCMC method requires the unnormalized posterior distribution of the parameters and a proposal distribution. In this thesis, we show how the posterior density function of the parameters of a state space model can be computed by filtering based methods, where the states are integrated out. This type of computation is then applied to estimate parameters of stochastic differential equations. Furthermore, we compute the partial derivatives of the log-posterior density function and use the hybrid Monte Carlo and scaled conjugate gradient methods to infer the parameters of stochastic differential equations. The computational efficiency of MCMC methods is highly depend on the chosen proposal distribution. A commonly used proposal distribution is Gaussian. In this kind of proposal, the covariance matrix must be well tuned. To tune it, adaptive MCMC methods can be used. In this thesis, we propose a new way of updating the covariance matrix using the variational Bayesian adaptive Kalman filter algorithm.

Distribution network design and optimization

The objective of this thesis is to examine distribution network designs and modeling practices and create a framework to identify best possible distribution network structure for the case company. The main research question therefore is: How to optimize case company’s distribution network in terms of customer needs and costs? Theory chapters introduce the basic building blocks of the distribution network design and needed calculation methods and models. Framework for the distribution network projects was created based on the theory and the case study was carried out by following the defined framework. Distribution network calculations were based on the company’s sales plan for the years 2014 - 2020. Main conclusions and recommendations were that the new Asian business strategy requires high investments in logistics and the first step is to open new satellite DC in China as soon as possible to support sales and second possible step is to open regional DC in Asia within 2 - 4 years.

Changes in cell shape and desmin intermediate filament distribution are associated with down-regulation of desmin expression in C2C12 myoblasts grown in the absence of extracellular Ca2+

Desmin is the main intermediate filament (IF) protein of muscle cells. In skeletal muscle, desmin IFs form a scaffold that interconnects the entire contractile apparatus with the subsarcolemmal cytoskeleton and cytoplasmic organelles. The interaction between desmin and the sarcolemma is mediated by a number of membrane proteins, many of which are Ca2+-sensitive. In the present study, we analyzed the effects of the Ca2+ chelator EGTA (1.75 mM) on the expression and distribution of desmin in C2C12 myoblasts grown in culture. We used indirect immunofluorescence microscopy and reverse transcription polymerase chain reaction (RT-PCR) to analyze desmin distribution and expression in C2C12 cells grown in the presence or absence of EGTA. Control C2C12 myoblasts showed a well-spread morphology after a few hours in culture and became bipolar when grown for 24 h in the presence of EGTA. Control C2C12 cells showed a dense network of desmin from the perinuclear region to the cell periphery, whereas EGTA-treated cells showed desmin aggregates in the cytoplasm. RT-PCR analysis revealed a down-regulation of desmin expression in EGTA-treated C2C12 cells compared to untreated cells. The present results suggest that extracellular Ca2+ availability plays a role in the regulation of desmin expression and in the spatial distribution of desmin IFs in myoblasts, and is involved in the generation and maintenance of myoblast cell shape.

Apolipoprotein E polymorphism distribution in an elderly Brazilian population: the Bambuí Health and Aging Study

Apolipoprotein E (ApoE) is one of the most extensively studied genes in the context of aging, but there are few population-based studies on ApoE polymorphism in the elderly in developing countries. The objective of the present study was to assess ApoE allele and genotype distribution in a large elderly community-based sample and its association with age, sex and skin color. Participants included 1408 subjects (80.8% of all residents aged ³60 years) residing in Bambuí city, MG, Brazil. The DNA samples were subjected to the polymerase chain reaction amplification, followed by the restriction fragment length polymorphism technique, with digestion by HhaI. Analysis was carried out taking into consideration the six ApoE genotypes (e3/e3, e3/e4, e2/e3, e4/e4, e2/e4, and e2/e2), the three ApoE alleles, and the number of ApoE4 alleles for each individual. The e3 allele predominated (80.0%), followed by e4 (13.5%) and e2 (6.5%). All six possible genotypes were observed, the e3/e3 genotype being the most frequent (63.4%). This distribution was similar to that described in other western populations. Sex was not associated with number of ApoE4 alleles. Black skin color was significantly and independently associated with the presence of two ApoE4 alleles (age-sex adjusted OR = 7.38; 95%CI = 1.93-28.25), showing that the African-Brazilian elderly have a high prevalence of the e4 allele, as observed in blacks from Africa. No association between number of ApoE4 alleles and age was found, suggesting the absence of association of ApoE genotype with mortality in this population.

Approximation de la distribution a posteriori d'un modèle Gamma-Poisson hiérarchique à effets mixtes

La méthode que nous présentons pour modéliser des données dites de "comptage" ou données de Poisson est basée sur la procédure nommée Modélisation multi-niveau et interactive de la régression de Poisson (PRIMM) développée par Christiansen et Morris (1997). Dans la méthode PRIMM, la régression de Poisson ne comprend que des effets fixes tandis que notre modèle intègre en plus des effets aléatoires. De même que Christiansen et Morris (1997), le modèle étudié consiste à faire de l'inférence basée sur des approximations analytiques des distributions a posteriori des paramètres, évitant ainsi d'utiliser des méthodes computationnelles comme les méthodes de Monte Carlo par chaînes de Markov (MCMC). Les approximations sont basées sur la méthode de Laplace et la théorie asymptotique liée à l'approximation normale pour les lois a posteriori. L'estimation des paramètres de la régression de Poisson est faite par la maximisation de leur densité a posteriori via l'algorithme de Newton-Raphson. Cette étude détermine également les deux premiers moments a posteriori des paramètres de la loi de Poisson dont la distribution a posteriori de chacun d'eux est approximativement une loi gamma. Des applications sur deux exemples de données ont permis de vérifier que ce modèle peut être considéré dans une certaine mesure comme une généralisation de la méthode PRIMM. En effet, le modèle s'applique aussi bien aux données de Poisson non stratifiées qu'aux données stratifiées; et dans ce dernier cas, il comporte non seulement des effets fixes mais aussi des effets aléatoires liés aux strates. Enfin, le modèle est appliqué aux données relatives à plusieurs types d'effets indésirables observés chez les participants d'un essai clinique impliquant un vaccin quadrivalent contre la rougeole, les oreillons, la rub\'eole et la varicelle. La régression de Poisson comprend l'effet fixe correspondant à la variable traitement/contrôle, ainsi que des effets aléatoires liés aux systèmes biologiques du corps humain auxquels sont attribués les effets indésirables considérés.

La distribution de films au Québec en 2011

En 2010, les écrans du Québec étaient occupés par des films américains dans une proportion de 78,4 %. Ce mémoire veut mettre en lumière les mécanismes de la distribution de films au Québec et la responsabilité des distributeurs dans l'offre cinématographique nationale. Après avoir décrit les cinq étapes de la fabrication d'un film, nous dressons l'inventaire des politiques fédérales et provinciales qui régissent la distribution au Québec. Nous établissons ensuite des parallèles entre les politiques nationales et celles de pays de l'Union européenne afin de considérer comment et pourquoi, pour protéger leur identité « de la déferlante américaine », ces pays tentent de réglementer l'industrie cinématographique. Par la suite, nous observons quel est le rôle de la distribution dans la filière économique en général et dans l'industrie québécoise du cinéma en particulier à travers une série d'entrevues semi-dirigées avec les principaux protagonistes du milieu. Nous nous intéressons subséquemment à l'avenir de la distribution cinématographique relativement à la dématérialisation des images et du son. Finalement, nous concluons que le menu cinématographique québécois est composé par le positionnement des distributeurs américains et québécois en réponse à une demande artificielle des consommateurs reflétée par la billetterie (box-office); positionnement légitimé par les politiques gouvernementales.

Polyamides and polyesters made of bile acids in the main chain

La préparation de polymères à base d’acides biliaires, molécules biologiques, a attiré l'attention des chercheurs en raison des applications potentielles dans les domaines biomédicaux et pharmaceutiques. L’objectif de ce travail est de synthétiser de nouveaux biopolymères dont la chaîne principale est constituée d’unités d’acides biliaires. La polymérisation par étapes a été adoptée dans ce projet afin de préparer les deux principales classes de polymères utilisés en fibres textiles: les polyamides et les polyesters. Des monomères hétéro-fonctionnels à base d’acides biliaires ont été synthétisés et utilisés afin de surmonter le déséquilibre stoechiométrique lors de la polymérisation par étapes. Le dérivé de l’acide lithocholique modifié par une fonction amine et un groupement carboxylique protégé a été polymérisé en masse à températures élevées. Les polyamides obtenus sont très peu solubles dans les solvants organiques. Des polyamides et des polyesters solubles en milieu organique ont pu être obtenus dans des conditions modérées en utilisant l’acide cholique modifié par des groupements azide et alcyne. La polymérisation a été réalisée par cycloaddition azoture-alcyne catalysée par l'intermédiaire du cuivre(Ι) avec deux systèmes catalytiques différents, le bromure de cuivre(I) et le sulfate de cuivre(II). Seul le bromure de cuivre(Ι) s’est avéré être un catalyseur efficace pour le système, permettant la préparation des polymères avec un degré de polymérisation égale à 50 et une distribution monomodale de masse moléculaire (PDI ˂ 1.7). Les polymères synthétisés à base d'acide cholique sont thermiquement stables (307 °C ≤ Td ≤ 372 °C) avec des températures de transition vitreuse élevées (137 °C ≤ Tg ≤ 167 °C) et modules de Young au-dessus de 280 MPa, dépendamment de la nature chimique du lien.

Droit du transport intermodal international de marchandises : une perspective « supply chain management ».

La thèse propose d’introduire une perspective globale dans le traitement juridique du transport intermodal international qui prendrait racine dans la stratégie logistique des entreprises. La conception juridique se heurte, en effet, aux évolutions opérationnelles et organisationnelles des transports et aboutit à une incertitude juridique. Les transporteurs ont dû s’adapter aux exigences d’optimisation des flux des chargeurs dont les modes de production et de distribution reposent sur le supply chain management (SCM). Ce concept est le fruit de la mondialisation et des technologies de l’information. La concurrence induite par la mondialisation et le pilotage optimal des flux ont impulsé de nouvelles stratégies de la part des entreprises qui tentent d’avoir un avantage concurrentiel sur le marché. Ces stratégies reposent sur l’intégration interfonctionnelle et interoganisationnelle. Dans cette chaîne logistique globale (ou SCM) l’intermodal est crucial. Il lie et coordonne les réseaux de production et de distribution spatialement désagrégés des entreprises et, répond aux exigences de maîtrise de l’espace et du temps, à moindre coût. Ainsi, le transporteur doit d’une part, intégrer les opérations de transport en optimisant les déplacements et, d’autre part, s’intégrer à la chaîne logistique du client en proposant des services de valeur ajoutée pour renforcer la compétitivité de la chaîne de valeur. Il en découle une unité technique et économique de la chaîne intermodale qui est pourtant, juridiquement fragmentée. Les Conventions internationales en vigueur ont été élaborées pour chaque mode de transport en faisant fi de l’interaction entre les modes et entre les opérateurs. L’intermodal est considéré comme une juxtaposition des modes et des régimes juridiques. Ce dépeçage juridique contraste avec la gestion de la chaîne intermodale dont les composantes individuelles s’effacent au profit de l’objectif global à atteindre. L’on expose d’abord l’ampleur de l’incertitude juridique due aux difficultés de circonscrire le champ d’opérations couvert par les Conventions en vigueur. Une attention est portée aux divergences d’interprétations qui débouchent sur la « désunification » du droit du transport. On s’intéresse ensuite aux interactions entre le transport et la chaîne logistique des chargeurs. Pour cela, on retrace l’évolution des modes de production et de distribution de ces derniers. C’est effectivement de la stratégie logistique que découle la conception de la chaîne intermodale. Partant de ce système, on identifie les caractéristiques fondamentales de l’intermodal. La thèse aboutit à dissiper les confusions liées à la qualification de l’intermodal et qui sont à la base des divergences d’interprétations et de l’incertitude juridique. De plus, elle met en exergue l’unité économique du contrat de transport intermodal qui devrait guider la fixation d’un régime de responsabilité dédié à ce système intégré de transport. Enfin, elle initie une approche ignorée des débats juridiques.

Detecting recombination in 4-taxa DNA sequence alignments with Bayesian hidden Markov models and Markov chain Monte Carlo

This article presents a statistical method for detecting recombination in DNA sequence alignments, which is based on combining two probabilistic graphical models: (1) a taxon graph (phylogenetic tree) representing the relationship between the taxa, and (2) a site graph (hidden Markov model) representing interactions between different sites in the DNA sequence alignments. We adopt a Bayesian approach and sample the parameters of the model from the posterior distribution with Markov chain Monte Carlo, using a Metropolis-Hastings and Gibbs-within-Gibbs scheme. The proposed method is tested on various synthetic and real-world DNA sequence alignments, and we compare its performance with the established detection methods RECPARS, PLATO, and TOPAL, as well as with two alternative parameter estimation schemes.

Intelligent wireless web services: context-aware computing in construction-logistics supply chain

The construction industry has incurred a considerable amount of waste as a result of poor logistics supply chain network management. Therefore, managing logistics in the construction industry is critical. An effective logistic system ensures delivery of the right products and services to the right players at the right time while minimising costs and rewarding all sectors based on value added to the supply chain. This paper reports on an on-going research study on the concept of context-aware services delivery in the construction project supply chain logistics. As part of the emerging wireless technologies, an Intelligent Wireless Web (IWW) using context-aware computing capability represents the next generation ICT application to construction-logistics management. This intelligent system has the potential of serving and improving the construction logistics through access to context-specific data, information and services. Existing mobile communication deployments in the construction industry rely on static modes of information delivery and do not take into account the worker’s changing context and dynamic project conditions. The major problems in these applications are lack of context-specificity in the distribution of information, services and other project resources, and lack of cohesion with the existing desktop based ICT infrastructure. The research works focus on identifying the context dimension such as user context, environmental context and project context, selection of technologies to capture context-parameters such wireless sensors and RFID, selection of supporting technologies such as wireless communication, Semantic Web, Web Services, agents, etc. The process of integration of Context-Aware Computing and Web-Services to facilitate the creation of intelligent collaboration environment for managing construction logistics will take into account all the necessary critical parameters such as storage, transportation, distribution, assembly, etc. within off and on-site project.

A detailed single chain model for molten poly(tetrafluoroethylene) from a novel structure refinement technique on neutron scattering data

We present a new methodology that couples neutron diffraction experiments over a wide Q range with single chain modelling in order to explore, in a quantitative manner, the intrachain organization of non-crystalline polymers. The technique is based on the assignment of parameters describing the chemical, geometric and conformational characteristics of the polymeric chain, and on the variation of these parameters to minimize the difference between the predicted and experimental diffraction patterns. The method is successfully applied to the study of molten poly(tetrafluoroethylene) at two different temperatures, and provides unambiguous information on the configuration of the chain and its degree of flexibility. From analysis of the experimental data a model is derived with CC and CF bond lengths of 1.58 and 1.36 Å, respectively, a backbone valence angle of 110° and a torsional angle distribution which is characterized by four isometric states, namely a split trans state at ± 18°, giving rise to a helical chain conformation, and two gauche states at ± 112°. The probability of trans conformers is 0.86 at T = 350°C, which decreases slightly to 0.84 at T = 400°C. Correspondingly, the chain segments are characterized by long all-trans sequences with random changes in sign, rather anisotropic in nature, which give rise to a rather stiff chain. We compare the results of this quantitative analysis of the experimental scattering data with the theoretical predictions of both force fields and molecular orbital conformation energy calculations.

Influence of the size distribution of particles on the viscous property of an electrorheological fluid

The influence of the size distribution of particles on the viscous property of an electrorheological fluid has been investigated by the molecular dynamic simulation method. The shear stress of the fluid is found to decrease with the increase of the variance sigma(2) of the Gaussian distribution of the particle size, and then reach a steady value when sigma is larger than 0.5. This phenomenon is attributed to the influence of the particle size distribution on the dynamic structural evolution in the fluid as well as the strength of the different chain-like structures formed by the particles.

Seafood safety and quality: an analysis of the supply chain in the Sultanate of Oman

The globalization of trade in fish has created many challenges for the developing world specifically with regard to food safety and quality. International organisations have established a good basis for standards in international trade. Whilst these requirements are frequently embraced by the major importers (such as Japan, the EU and the USA), they often impose additional safety requirements and regularly identify batches which fail to meet their strict standards. Creating an effective national seafood control system which meets both the internal national needs as well the requirements for the export market can be challenging. Many countries adopt a dual system where seafood products for the major export markets are subject to tight control whilst the majority of the products (whether for the local market or for more regional trade) are less tightly controlled. With regional liberalization also occurring, deciding on appropriate controls is complex. In the Sultanate of Oman, fisheries production is one of the countries' chief sources of economic revenue after oil production and is a major source of the national food supply. In this paper the structure of the fish supply chain has been analysed and highlighted the different routes operating for the different markets. Although much of the fish are consumed within Oman, there is a major export trade to the local regional markets. Much smaller quantities meet the more stringent standards imposed by the major importing countries and exports to these are limited. The paper has considered the development of the Omani fish control system including the key legislative documents and the administrative structures that have been developed. Establishing modern controls which satisfy the demands of the major importers is possible but places additional costs on businesses. Enhanced controls such as HACCP and other management standards are required but can be difficult to justify when alternative markets do not specify these. These enhanced controls do however provide additional consumer protection and can bring benefits to local consumers. The Omani government is attempting to upgrade the system of controls and has made tremendous progress toward the implementation of HACCP and introducing enhanced management systems into its industrial sector. The existence of strengthened legislative and government support, including subsidies, has encouraged some businesses to implement HACCP. The current control systems have been reviewed and a SWOT analysis approach used to identify key factors for their future development. The study shows that seafood products in the supply chain are often exposed to lengthy handling and distribution process before reaching the consumers, a typical issue faced by many developing countries. As seafood products are often perishable, they safety is compromised if not adequately controlled. The enforcement of current food safety laws in the Sultanate of Oman is shared across various government agencies. Consequently, there is a need to harmonize all regulatory requirements, enhancing the domestic food protection and to continue to work towards a fully risk-based approach in order to compete successfully in the global market.

Controlled variation of monomer sequence-distribution in the synthesis of aromatic poly(ether ketone)s

The effects of varying the alkali metal cation in the high-temperature nucleophilic synthesis of a semi-crystalline, aromatic poly(ether ketone) have been systematically investigated, and striking variations in the sequence-distributions and thermal characteristics of the resulting polymers were found. Polycondensation of 4,4'-dihydroxybenzophenone with 1,3-bis(4-fluorobenzoyl)benzene in diphenylsulfone as solvent, in the presence of an alkali metal carbonate M2CO3 (M= Li, Na, K, or Rb) as base, affords a range of different polymers that vary in the distribution pattern of 2-ring and 3-ring monomer units along the chain. Lithium carbonate gives an essentially alternating and highly crystalline polymer, but the degree of sequence-randomisation increases progressively as the alkali metal series is descended, with rubidium carbonate giving a fully random and non-thermally-crystallisable polymer. Randomisation during polycondensation is shown to result from reversible cleavage of the ether linkages in the polymer by fluoride ions, and an isolated sample of alternating-sequence polymer is thus converted to a fully randomised material on heating with rubidium fluoride.

Expression Profile and Distribution of Efhc1 Gene Transcript During Rodent Brain Development

One of the putative causative genes for juvenile myoclonic epilepsy (JME) is EFHC1. We report here the expression profile and distribution of Efhc1 messenger RNA (mRNA) during mouse and rat brain development. Real-time polymerase chain reaction revealed that there is no difference in the expression of Efhc1 mRNA between right and left hemispheres in both species. In addition, the highest levels of Efhc1 mRNA were found at intra-uterine stages in mouse and in adulthood in rat. In common, there was a progressive decrease in Efhc1 expression from 1-day-old neonates to 14-day-old animals in both species. In situ hybridization studies showed that rat and mouse Efhc1 mRNAs are expressed in ependymal cells of ventricle walls. Our findings suggest that Efhc1 expression is more important during initial phases of brain development and that at this stage it could be involved in key developmental mechanisms underlying JME.