[pt] Este estudo explora fontes de dados não convencionais como dados textuais de jornais e pesquisas de internet do Google Trends em dois problemas empíricos: (i) analisar o impacto da mobilidade sobre o número de casos e mortes por Covid-19; (ii) nowcasting do PIB em alta-frequência. O primeiro artigo usa fontes de dados não estruturados como controle para fatores comportamentais não observados e encontra que um aumento na mobilidade residencial diminui significativamente o número de casos e mortes num horizonte de quatro semanas. O segundo artigo usa fontes de dados não estruturadas para fazer um nowcasting semanal do PIB, mostrando que dados textuais e Google Trends pode aumentar a qualidade das projeções (medido pelo EQM, EAM e outras métricas) comparado com as expectativas de mercado do Focus como base. Em ambos casos, dados não estruturados reveleram-se fontes ricas de informação não codificadas em indicadores estruturados convencionais. / [en] This study exploits non-conventional data sources such as newspaper textual data and internet searches from Google Trends in two empirical problems: (i) analysing the impacts of mobility on cases and deaths due to Covid-19; (ii) nowcasting GDP in high-frequency. The first paper resorts to unstructured data to control for non-observable behavioural effects and finds that an increase in residential mobility significantly reduces Covid-19 cases and deaths over a 4-week horizon. The second paper uses unstructured data sources to nowcast GDP on a weekly basis, showing that textual data and Google Trends can significantly enhance the quality of nowcasts (measured by MSE, MAE and other metrics) compared to Focus s market expectations as a benchmark. In both cases, unstructured data was revealed to be a valuable source of information not encoded in structured indicators.

Thesis (PhD)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: With the depletion of high quality conventional materials for road pavements, the consideration of cement stabilisation for sub-base and base layer materials often provide a feasible solution to the road industry. Like all pavement material types, the design inputs should be determined using reliable test methods, which provide a good indication of the property of materials. Any evaluation should provide a better understanding of the engineering and behavioural properties of the materials. This should form the basis for ascertaining their suitability for use in the pavement structure. However, the road industry is dependent on strength testing of cement-stabilised materials as a means to ascertain material suitability for use. Strength alone does not offer reliable insight regarding the performance and durability of the stabilised layer. This is because a cement-stabilised layer may be very stiff but not strong enough to withstand the loading and endure adverse environmental conditions. Similarly, the stabilised layer may be prone to cracking emanating from shrinkage, which leads to performance and durability related distresses. A stabilised sub-base and base of the pavement structure experiences tensile stresses and strains under traffic loading. At laboratory level, the flexural beam test simulates to an acceptable degree the mode of stress to strain to which the pavement layer experiences. However, the test lacks a standard test protocol. This leads to inconsistencies while evaluating the same material type. Due to this fact, the formulation of a standard laboratory test procedure is necessary. Shrinkage cracking is one of the major causes of pavement failure. The manifestation of wide cracks leads to performance related distresses. Cracks provide zones for the infiltration of water into the underlying layers, an aspect that results in further deterioration of the pavement structure. However, the evaluation of shrinkage at laboratory level is not usually undertaken. Disregarding shrinkage evaluation stems from the fact that a number of guidelines consider it as a natural material characteristic. The road industry frequently depends on the use of low cement contents among other techniques as a means to mitigate shrinkage cracking in cement-stabilised layers. The selection of a mitigation measure usually lacks reliable data concerning the material’s shrinkage potential. Because of this, the requirement to evaluate shrinkage at laboratory level as part of a material property measure provides a good indication regarding the quality of material. Nanotechnology products such as the Nanotterra Soil® a polymer cement additive are purported to mitigate shrinkage cracking in cement-stabilised layers. However, their suitability for use remains unspecified and dependent on the stakeholders. With the development of a shrinkage method, the evaluation of shrinkage reducing products can be undertaken. This research proposes a flexural beam test protocol for cement-stabilised materials, comprising of a span-depth ratio of nine or greater as fitting to provide a reliable measure of the material’s flexural strength and elastic modulus. The developed shrinkage test method provides a good repeatability and is user friendly. The test provides a good indication of the shrinkage criteria of ferricrete and hornfels with and without the polymer. The efficacy of the polymer is dependent on the cement content in the mix and the type and quality of the material. The research provides insight pertaining to the characterisation of shrinkage, behavioural properties, and durability of cement-stabilised materials. Analysis of the shrinkage crack pattern reveals that use of the polymer lessens the development of tensile stress in a cement-stabilised layer. Equally, the application of the low cement contents for stabilisation may not result in cracking of the stabilised layer. This research contributes to a better understanding of cement-stabilised materials. / AFRIKAANSE OPSOMMING: Namate hoë kwaliteit konvensionele materiale uitgeput raak, word sementstabilisasie van stutlaag en kroonlaag materiale al hoe meer oorweeg en is dit ʼn geskikte oplossing vir die padbou-nywerheid. Soos vir alle padboumateriale moet die ontwerpeienskappe bepaal word deur middel van betroubare toetsmetodes wat ʼn goeie aanduiding van die materiaal se eienskappe sal gee. Enige evaluering moet ʼn beter insae in die materiaal se ingenieurseienskappe en gedrag oplewer. Dit moet dan die basis vorm om die materiaal se gebruik in ʼn padstruktuur te evalueer. Die padbou-nywerheidmaak grootliks staat op die toetsing van skuifsterkte van sementgestabiliseerde materiaal om die geskiktheid daarvan vir gebruik te bepaal. Sterkte op sigself lewer egter nie ʼn betroubare maatstaf van die materiaal se gedrag en duursaamheid nie. Dit is aangesien ʼn sementgestabiliseerde laag baie solied mag wees maar nogtans nie sterk genoeg om belasting te weerstaan en bestand teen omgewingstoestande te wees nie. Net so mag ʼn gestabiliseerde laag vatbaar vir kraakvorming as gevolg van krimping wees en dit kan lei tot duursaamheid-verwante en werkverwante skade. ʼn Gestabiliseerde stutlaag en kroonlaag in die plaveiselstruktuur is onderhewig aan trekspannings en vervormings as gevolg van verkeerslaste. Op laboratoriumvlak boots die balkbuigtoets die spanning en vervorming wat ʼn plaveisellaag ondervind tot ʼn aanvaarbaar hoë mate na. Die toets beskik nie oor ʼn standaard-toetsprosedure nie. Dit lei tot afwykings terwyl dieselfde materiaal evalueer word. Om hierdie rede is die ontwikkeling van ʼn standaard-laboratoriumprosedure nodig. Krimpkraking is een van die grootste oorsake van plaveiselswigting. Die onwikkeling van wye krake lei tot werksverwante skade. Krake veroorsaak areas vir die infiltrasie van water in die onderliggende plaveisellae wat verdere agteruitgang van die plaveiselstruktuur veroorsaak. Desnieteenstaande word ʼn evaluering van kraking op laboratoriumvlak selde gedoen. Dit spruit uit die feit dat ʼn aantal ontwerp-riglyne kraking as ʼn natuurlike materiaaleienskap beskou. Die padbounywerheid moet dikwels staatmaak, op onder andere, ʼn lae sementinhoud om krimpkraking te minimeer. Hierdie tipe benadering gaan dikwels mank aan betroubare inligting oor die materiaal se krimpingspotensiaal. Om hierdie rede is die ondersoek van krimping op laboratoriumvlak nodig as deel van die ondersoek van die materiaaleienskappe om die kwaliteit van materiale te bepaal. Minimeringstegnieke verander deurlopend. Die toepassing van nanotegnologieprodukte, soos Nanotterra Soil®, ‘n polimeersement bymiddel, wat na bewering krimpkraking in sementgestabiliseerde lae kan minimeer, kom voortdurend op die mark. Nogtans bly hulle geskiktheid ongespesifiseerd en afhanklik van die leweransiers. Indien ʼn krimptoetsmetode ontwikkel word, sal die effektiwiteit van krimpverminderingsmiddels getoets kan word. Hierdie navorsing stel die ontwikkeling van ʼn toetsprosedure vir ʼn balkbuigtoets voor met ʼn spanlengte tot diepteverhouding van minstens nege as betroubare maatstaf van ʼn materiaal se buigsterkte en modulus van elastisiteit. Die ontwikkelde krimptoetsmetode lewer ʼn goeie herhaalbaarheid en is gebruikersvriendelik. Die toets verskaf ʼn goeie aanduiding van krimpingskriteria van ferrikreet en horingfels met en sonder polimeer. Die effektiwiteit van die polimeer hang af van die sementinhoud in die mengsel asook die tipe en kwaliteit van die materiaal. Die navorsing verskaf insig aangaande die karakterisering van krimping, gedragseienskappe en duursaamheid van sementgestabiliseerde materiale. Die navorsing help mee om sementgestabiliseerde materiale beter te verstaan.

<p>Pharmacometric modelling is an increasingly used method for analysing the outcome from clinical trials in drug development. The model building process is complex and involves testing, evaluating and diagnosing a range of plausible models aiming to make an adequate inference from the observed data and predictions for future studies and therapy. </p><p>The aim of this thesis was to advance the approaches used in pharmacometrics by introducing improved models and methods for application in essential parts of model building procedure: (i) structural model development, (ii) stochastic model development and (iii) model diagnostics. </p><p>As a contribution to the structural model development, a novel flexible structural model for drug absorption, a transit compartment model, was introduced and evaluated. This model is capable of describing various drug absorption profiles and yet simple enough to be estimable from data available from a typical trial. As a contribution to the stochastic model development, three novel methods for parameter distribution estimation were developed and evaluated; a default NONMEM nonparametric method, an extended grid method and a semiparametric method with estimated shape parameters. All these methods are useful in circumstances when standard assumptions of parameter distributions in the population do not hold. The new methods provide less biased parameter estimates, better description of variability and better simulation properties of the model. As a contribution to model diagnostics, the most commonly used diagnostics were evaluated for their usefulness. In particular, diagnostics based on individual parameter estimates were systematically investigated and circumstances which are likely to misguide modelers towards making erroneous decisions in model development, relating to choice of structural, covariate and stochastic model components were identified. </p><p>In conclusion, novel approaches, insights and models have been provided to the pharmacometrics community. </p><p>Implementation of these advances to make model building more efficient and robust has been facilitated by development of diagnostic tools and automated routines.</p>

No description available.

Les ouvrages du génie civil ont une durée de vie limitée, et plusieurs d’entre eux nécessitent aujourd’hui une réparation, voire une reconstruction partielle. Ainsi le domaine des réparations et des renforcements en béton connaît depuis une dizaine d’années, une croissance importante dans le secteur du bâtiment et des travaux publics. Plusieurs études ont montré que le retrait et le fluage peuvent mener à de sévères problèmes liés à la durabilité. En effet, ces phénomènes peuvent être à l’origine de la fissuration, et même plus rarement de la ruine de l’ouvrage. Cette thèse a pour objectif d’étudier le comportement au retrait et au fluage d’éléments renforcés. Ce travail est divisé en deux parties principales. Dans la première partie, nous effectuons une étude expérimentale sur des éléments renforcés. Nous nous intéressons, dans la deuxième partie,à la modélisation du séchage afin de prévoir les transferts hydriques au sein du matériau, et le retrait de celui-ci. L’étude expérimentale vise à étudier l’effet du renforcement sur le comportement différé d’éprouvettes en mortier normalisé. L’approche théorique et numérique permet d’analyser les transferts hydriques et la diffusion de l’humidité au sein du matériau, et de calculer ainsi les déformations différées. Les essais expérimentaux montrent que le renforcement avec des matériaux composites réduit considérablement la déformation au fluage, mais n’apporte pas d’amélioration quant au retrait de séchage. / The works of civil engineering have one limited life span, and several of them require a repair today, even a partial rebuilding. Thus the field of the concrete repairs and strengthening knowssince ten years, a considerable growth in the sector of the building and public works. Several studies showed that the shrinkage and creep can lead to severe problems involved indurability. Indeed, these phenomena can be at the origin of cracking, and even more rarely of the ruin of the work. This thesis aims to study the shrinkage and creep behaviour of strengthened elements. This work is divided into two principal parts. In the first part, we carry out an experimental study on strengthened elements. We are interested, in the second part, to the modelling of drying in orderto envisage the hydrous transfers within material, and the shrinkage of this one. In the experimental study, we are interested to the strengthening effect on the differed behaviour of mortar specimens. The theoretical and numerical approach allows to analyze the hydrous transfers and the diffusion of humidity within material, and thus to calculate the differed deformations. The experimental tests show that the strengthening with composite materials reduced considerably the creep strain, but does not make improvement as for the shrinkage ofdrying.

The problem was to determine the effectiveness of using polyethylene glycol 1000 in the treatment of green wood for the purpose of forming projects made of wood veneer and of simple design for use in junior high or high school woodworking classes. The purpose of this study was to seek answers to the following questions. 1. Is polyethylene glycol 1000 an effective stabilizing agent for green wood veneer that can be used in school woodworking classes? 2. Can green wood veneer treated with polyethylene glycol 1000 be bent to form simple woodwork projects? 3. Can green wood veneer treated with polyethylene glycol 1000 be successfully used in junior high and high school woodworking classes? 4. What length of treatment time is best for green wood veneer that is to be used to form simple bent wood projects? 5. Is one-fourth inch thickness suitable for green wood veneer that is to be treated with polyethylene glycol 1000 and used to form simple bent wood projects?

Friction Stir Welding (FSW) was used to perform Bead on Plate (BOP) welds on 6061-T6511 aluminum extrusions. Using a DOE approach, tool rotation speed, clamp spacing, and clamping force were altered to ascertain their effects on distortion in the welded panels. Mechanical forces were monitored during the weld process. Both linear and out of plane distortion were measured on the welded extrusions. The Vickers hardness of the weld nugget was measured. The effect of each parameter on weld distortion was discovered and the mechanism of this link was suggested.

We provide a flexible means of estimating time-varying parameter models in a Bayesian framework. By specifying the state innovations to be characterized trough a threshold process that is driven by the absolute size of parameter changes, our model detects at each point in time whether a given regression coefficient is con stant or time-varying. Moreover, our framework accounts for model uncertainty in a data-based fashion through Bayesian shrinkage priors on the initial values of the states. In a simulation, we show that our model reliably identifies regime shifts in cases where the data generating processes display high, moderate, and low numbers of movements in the regression parameters. Finally, we illustrate the merits of our approach by means of two applications. In the first application we forecast the US equity premium and in the second application we investigate the macroeconomic effects of a US monetary policy shock. / Series: Research Report Series / Department of Statistics and Mathematics

O objetivo deste trabalho é controlar a incidência de fissuras em fibrocimentos aplicando o conceito de materiais com gradação funcional através da protensão química gerada pela aplicação de silicato de sódio alcalino entre as camadas dos fibrocimentos. Atualmente é bastante comum os fibrocimentos apresentarem fissuras ao longo das bordas devido aos gradientes de umidade gerados durante a estocagem das pilhas de telhas no pátio das indústrias. O potencial da protensão química foi avaliado através da porosidade total, da quantificação das fases hidratadas, da retração por secagem e do desempenho mecânico. A aplicação de silicato de sódio alcalino no ligante CPII F provocou retração por secagem 1,5 vezes maior que a referência aos 91 dias. O módulo de ruptura (MOR) não sofreu alteração, mas o limite de proporcionalidade da matriz (LOP) aumentou em torno de 95%. O módulo de elasticidade dinâmico foi 13 % maior. O aumento da retração por secagem e o ganho de desempenho mecânico apontam o potencial da protensão química para o controle da fissuração em fibrocimentos produzidos pelo processo Hatschek. / The objective of this work is controlling the incidence of cracks in fiber cement, using the concept of functionally graded materials through the chemical prestressing, generated by application of alkaline sodium silicate among fiber cement layers. Currently, its very common the fiber cements present cracks along the edges due to moisture gradients, caused during storage of piles of tiles in the courtyard of the industry. The chemical prestressing potential was evaluated through of the total porosity, the quantification of hydrate phases, the drying shrinkage and the mechanical performance. The application of alkaline sodium silicate in the cement CPII F caused drying shrinkage 1,5 times greater than the reference to 91 days. The modulus of rupture (MOR) didnt suffer change, but the proportional limit of matrix (LOP) increased by around 95%. The dynamic modulus of elasticity was 13% higher. The increase of drying shrinkage and the mechanical performance gain indicate the chemical prestressing potential to control the cracking in fiber cement produced by the process Hatschek.

Proposição: O objetivo deste estudo foi avaliar como a contração volumétrica (CV), o grau de conversão (GC), a taxa máxima de polimerização (RPmax) e a tensão de polimerização (TPmax) de dois compósitos experimentais são influenciados pela concentração de fotoiniciadores na matriz e pela irradiância aplicada na fotopolimerização. Material e métodos: Duas séries de monômeros foram formuladas, com partes iguais em peso dos seguintes monômeros: série B= Bis-GMA/ TEGDMA; série U= Bis-GMA/ UDMA / TEGDMA. Essas séries foram combinadas a três níveis de fotoiniciadores, num total de seis materiais experimentais, com a concentração em porcentagem de peso de uma amina terciária (EDMAB) e canforoquinona (CQ), respectivamente: alta (H)= 0,8 / 1,6; intermediária (M)= 0,4 / 0,8 e baixa (L)= 0,2 / 0,4. A todas as misturas, 75% e peso de vidro de estrôncio e 5% em peso de OX-50 (sílica coloidal) foram adicionados. O estudo foi dividido em três módulos: no primeiro, os seis compósitos foram testados quanto aos parâmetros descritos acima para a mesma dose de energia aplicada. No segundo, os mesmos materiais foram testados com a dose de energia ajustada de acordo com a concentração de fotoiniciadores de modo que todos os níveis atingissem graus de conversão semelhantes. No terceiro módulo, os materiais com a concentração \"M\" de fotoiniciadores foram submetidos a uma de três irradiâncias: 200, 400 ou 600 mW/cm2, com o tempo de exposição ajustado para que a dose de energia fosse mantida constante. CV foi registrada em um dilatômetro de mercúrio, GC foi avaliado através de FTIR, RPmax foi obtida através de um DSC 7 e a TP foi avaliada em um aparato de compliance controlado, descrito por Sakaguchi et al., 2004. Resultados: Em todos os módulos, não houve interação entre os fatores e assim, as médias foram agrupadas por série de monômeros e nível de fotoiniciadores (módulos 1 e 2) ou nível de irradiância (módulo 3). No módulo 1, as concentrações \"H\" e \"M\" apresentaram GC semelhante, ambas estatisticamente superiores à apresentada pelo grupo com o menor conteúdo de CQ/amina. A série U apresentou GC superior à da série B. RPmax aumentou significantemente com a concentração de fotoiniciadores entre cada um dos níveis estudados (p<0,001), e foi maior para a série U (p<0,05). O grupo de concentração \"H\" apresentou valores de TPmax maiores do que os grupos de concentração \"L\", enquanto que \"M\" apresentou valores de TPmax semelhantes aos dos outros dois (p<0,001). As séries B e U apresentaram TPmax equivalente (p>0,05). CV foi maior para \"L\", estatisticamente diferente das outras duas. A série B apresentou maior CV (p<0,05 para ambos os fatores). No módulo 2, o GC não foi influenciado pela concentração de fotoiniciadores (p>0,05). A série U apresentou maior GC comparado com a série B (p<0,001). RPmax aumentou com a concentração de fotoiniciadores entre cada um dos níveis estudados e foi maior para a série U (p<0,001 para ambos). TPmax não foi influenciada pela concentração de fotoiniciadores (p>0,05). A série B apresentou maior TPmax comparado com a série U (p<0,001). CV foi maior para \"L\", estatisticamente diferente das outras duas. A série B apresentou maiores valores de CV (p<0,001 para ambos os fatores). No módulo 3, GC variou significantemente apenas com a irradiância (p<0,001), sendo que os grupos de alta irradiância apresentaram valores estatisticamente mais baixos comparados aos níveis de irradiância médio e baixo. RPmax, aumentou com a irradiância e foi maior para a série U (p<0,001 para ambos). A série B apresentou valores de TPmax do que os da série U. CV não foi influenciada pela irradiância (p=0,442). A série B apresentou CV estatisticamente maior que a série U (p<0,001). Conclusões: De maneira geral, podemos dizer que CV, GC e RPmax mostraram uma interação bastante complexa e, nas condições deste estudo, não foi possível determinar a contribuição relativa de cada um destes fatores no desenvolvimento da tensão de polimerização. No entanto, parece haver uma tendência de TPmax ser influenciada mais pronunciadamente pelo GC ou CV do que pela RPmax, como demonstrado nos módulos 2 e 3. / Proposition: The objective of this study was to evaluate how the volumetric shrinkage (VS), degree of conversion (DC), maximum rate of reaction (RPmax), polymerization stress (PSmax) and maximum rate of stress development (RSmax) of two experimental composites are influenced by the photoinitiator concentration ([photo]) and by the irradiance. Methods and materials: Bis-GMA/TEGDMA (B series) and Bis GMA/UDMA/TEGDMA (U series) were mixed in equal parts in weight. A tertiary amine (EDMAB) and camphoroquinone (CQ), respectively, were added in three concentrations: high (H)= 0.8/1.6; intermediate (M)= 0.4/0.8 and low (L)= 0.2/0.4 (in wt%). In all mixtures, 80 wt% filler was added. The study was divided in three sections: I) composites were photoactivated with 13 J/cm2. II) radiant exposure was adjusted so that all [photo] would achieve similar DC. III) [photo] \"M\" materials were submitted to three irradiances: 200 (L), 400 (I) or 600 (H) mW/cm2, with exposure time adjusted to convey the same radiant exposure. VS was registered in a mercury dilatometer, DC was evaluated by FTIR spectroscopy, RPmax was obtained in a DSC 7 and PSmax was evaluated in a controlled compliance device, described previously (SAKAGUCHI; WILTBANK; MURCHISON, 2004a). Results: Because no interaction between the factors was observed (except for RSmax in sections I and III), means were pooled for monomer series and [photo] (sections I and II) or irradiance (section III). I) concentrations \"H\" and \"M\" presented similar DC, both statistically superior to \"L\" (p<0.001). U series presented higher DC compared to B series. RPmax increased significantly with photoinitiator concentration between each of the studied levels (p<0.001), and was higher for U series (p<0.05). VS was higher for \"L\", statistically different than the other two (p<0.05). B series presented higher VS (p<0.05). The group with \"H\" concentration presented PSmax higher than \"L\", while \"M\" presented values similar to both (p<0.001). B and U series presented equivalent PSmax (p=0.284). For B series, RSmax increased 100% between the lower e intermediate [photo]. For U series, this increase was only around 50%. II) DC was not influenced by [photo] (p=0.388). U series presented higher DC compared to B series (p<0.001). RPmax increased with [photo] between all studied levels and was higher for U series (p<0.001 for both). VS was higher for \"L\", statistically different from the other two (p<0.001). B series presented higher VS (p<0.001). PSmax was not influenced by the photoinitiator concentration (p=0119). B series presented higher PSmax compared to U series (p<0.001). RSmax increased with [photo] and was higher for U series (p<0.001). III) High irradiance groups presented statistically lower DC compared to the other irradiance levels (p<0.001). Monomer series did not influence DC (p=0.793). RPmax increased with the irradiance and was higher for U series (p<0.001 for both). VS was not influenced by the irradiance (p=0.442). B series presented VS statistically higher than U series (p<0.001). \"H\" groups presented higher PSmax compared to \"L\". \"M\" presented PSmax similar to both (p<0.001). B series presented PSmax higher than U series. While no difference in RSmax between irradiance leves was observed for B series, RSmax increased with the irradiance for U series. Conclusions: There seems to be a trend for PSmax to be influenced more markedly by DC or VS than by RPmax. However, VS, DC and RPmax showed a quite complex interaction and, in the conditions of the present study, it was not possible to determine the relative contribution of each of these factors on stress development.