Le travail général de cette thèse consiste à étendre les outils analytiques et algébriques usuellement employés dans la résolution de problèmes combinatoires déterministes à un cadre combinatoire stochastique. Deux cadres distincts sont étudiés : les problèmes combinatoires stochastiques discrets et les problèmes stochastiques continus. Le cadre discret est abordé à travers le problème de la forêt couvrante de poids maximal dans une formulation Two-Stage à multi-scénarios. La version déterministe très connue de ce problème établit des liens entre la fonction de rang dans un matroïde et la formulation duale, via l'algorithme glouton. La formulation stochastique discrète du problème de la forêt maximale couvrante est transformée en un problème déterministe équivalent, mais du fait de la multiplicité des scénarios, le dual associé est en quelque sorte incomplet. Le travail réalisé ici consiste à comprendre en quelles circonstances la formulation duale atteint néanmoins un minimum égal au problème primal intégral. D'ordinaire, une approche combinatoire classique des problèmes de graphes pondérés consiste à rechercher des configurations particulières au sein des graphes, comme les circuits, et à explorer d'éventuelles recombinaisons. Pour donner une illustration simple, si on change d'une manière infinitésimale les valeurs de poids des arêtes d'un graphe, il est possible que la forêt couvrante de poids maximal se réorganise complètement. Ceci est vu comme un obstacle dans une approche purement combinatoire. Pourtant, certaines grandeurs analytiques vont varier de manière continue en fonction de ces variations infinitésimales, comme la somme des poids des arêtes choisies. Nous introduisons des fonctions qui rendent compte de ces variations continues, et nous examinons dans quels cas les formulations duales atteignent la même valeur que les formulations primales intégrales. Nous proposons une méthode d'approximation dans le cas contraire et nous statuons sur la NP complétude de ce type de problème.Les problèmes stochastiques continus sont abordés via le problème de sac à dos avec contrainte stochastique. La formulation est de type ``chance constraint'', et la dualisation par variable lagrangienne est adaptée à une situation où la probabilité de respecter la contrainte doit rester proche de $1$. Le modèle étudié est celui d'un sac à dos où les objets ont une valeur et un poids déterminés par des distributions normales. Dans notre approche, nous nous attachons à appliquer des méthodes de gradient directement sur la formulation en espérance de la fonction objectif et de la contrainte. Nous délaissons donc une possible reformulation classique du problème sous forme géométrique pour détailler les conditions de convergence de la méthode du gradient stochastique. Cette partie est illustrée par des tests numériques de comparaison avec la méthode SOCP sur des instances combinatoires avec méthode de Branch and Bound, et sur des instances relaxées. / The global purpose of this thesis is to study the conditions to extend analytical and algebraical properties commonly observed in the resolution of deterministic combinatorial problems to the corresponding stochastic formulations of these problems. Two distinct situations are treated : discrete combinatorial stochastic problems and continuous stochastic problems. Discrete situation is examined with the Two Stage formulation of the Maximum Weight Covering Forest. The well known corresponding deterministic formulation shows the connexions between the rank function of a matroid, the greedy algorithm , and the dual formulation. The discrete stochastic formulation of the Maximal Covering Forest is turned into a deterministic equivalent formulation, but, due to the number of scenarios, the associated dual is not complete. The work of this thesis leads to understand in which cases the dual formulation still has the same value as the primal integer formulation. Usually, classical combinatorial approaches aim to find particular configurations in the graph, as circuits, in order to handle possible reconfigurations. For example, slight modifications of the weights of the edges might change considerably the configuration of the Maximum Weight Covering Forest. This can be seen as an obstacle to handle pure combinatorial proofs. However, some global relevant quantities, like the global weight of the selected edges during the greedy algorithm, have a continuous variation in function of slight modifications. We introduce some functions in order to outline these continuous variations. And we state in which cases Primal integral problems have the same objective values as dual formulations. When it is not the case, we propose an approximation method and we examine the NP completeness of this problem.Continuous stochastic problems are presented with the stochastic Knapsack with chance constraint. Chance constraint and dual Lagrangian formulation are adapted in the case where the expected probability of not exceeding the knapsack capacity is close to $1$. The introduced model consists in items whose costs and rewards follow normal distributions. In our case, we try to apply direct gradient methods without reformulating the problem into geometrical terms. We detail convergence conditions of gradient based methods directly on the initial formulation. This part is illustrated with numerical tests on combinatorial instances and Branch and Bound evaluations on relaxed formulations.

Wheat-based thin stillage (W-TS) is liquid by-product of wheat ethanol production and contains valuable chemical intermediates such as 1,3-propanediol (1,3-PD), acetic acid, and glycerophosphorylcholine. Unfortunately, these compounds cannot be recovered/extracted easily due to the presence of high boiling point and hygroscopic solutes and unfermented particles from ethanol fermentation. Fermentation improvement study using endemic bacteria augmented with Lactobacillus panis PM1B discovered that glucose, incubation temperature, micronutrients, and pH adjustment affected two-stage fermentation (TSF). Importantly, fermentation could be scaled to a 210 L fermenter where 2% (w/v) 1,3-PD was produced. Unfermented particles should be removed prior to compound recovery. TSF was effective in producing solutions that were virtually free of colloids. Bacteria present in TSF system produced anoxic gas and exopolysaccharides and the combined action produced substantially clear solution. On the other hand, recovered particles, rich in lactobacilli, had a high protein content (50%, w/w, dry basis), which might be useful as an animal feed ingredient. Washing processes could lower moisture content and recover a high protein slurry (60% w/w, dry basis). Practical processes that incorporated fermentation using Lactobacilli could add substantial value to thin stillage and increase the value of products from ethanol production. These processes are scalable and readily implemented.

This paper develops an estimator for higher-order spatial autoregressive panel data error component models with spatial autoregressive disturbances, SARAR(R,S). We derive the moment conditions and optimal weighting matrix without distributional assumptions for a generalized moments (GM) estimation procedure of the spatial autoregressive parameters of the disturbance process and define a generalized two-stage least squares estimator for the regression parameters of the model. We prove consistency of the proposed estimators, derive their joint asymptotic distribution, and provide Monte Carlo evidence on their small sample performance.

This paper develops a unified framework for fixed and random effects estimation of higher-order spatial autoregressive panel data models with spatial autoregressive disturbances and heteroskedasticity of unknown form in the idiosyncratic error component. We derive the moment conditions and optimal weighting matrix without distributional assumptions for a generalized moments (GM) estimation procedure of the spatial autoregressive parameters of the disturbance process and define both a random effects and a fixed effects spatial generalized two-stage least squares estimator for the regression parameters of the model. We prove consistency of the proposed estimators and derive their joint asymptotic distribution, which is robust to heteroskedasticity of unknown form in the idiosyncratic error component. Finally, we derive a robust Hausman-test of the spatial random against the spatial fixed effects model. (authors' abstract) / Series: Department of Economics Working Paper Series

This thesis complements current studies by focusing on developed OECD countries as they are the major sources and recipients of world FDI and current studies relating to developed countries using aggregate country FDI data are limited. This study empirically tests the determinants of FDI inflows and outflows and their relationship with economic growth using 2SLS simultaneous equations model between 1981 and 2008 for a sample of 20 developed OECD countries. The empirical findings suggest that FDI inflows do not contribute to economic growth in the host country and economic growth positively affects FDI inflows. In addition, trade openness and flexible employment protection legislation in the host country attract FDI inflows. In terms of FDI outflows, the results show that FDI outflows reduce economic growth in the home country, while economic growth in the home country increases FDI outflows. Moreover, high past level of outward FDI stock, trade openness, low labour cost and currency depreciation in the home country provide incentives for domestic firms to invest abroad. Therefore, this study does not support offering special incentives to foreign investors to attract FDI inflows or offering promotional policies to domestic firms to encourage FDI outflows. Instead, government should provide incentives for domestic investment and other sound policies to increase economic growth, which in itself provides a good environment to attract FDI inflows and to encourage FDI outflows. Keywords: FDI inflows, FDI outflows, two stage least squares simultaneous equations, economic growth, labour market flexibility.

Foram operados dois reatores UASB de bancada (volumes de 39,0 e 10,5 L) instalados em série, alimentados com águas residuárias de suinocultura com concentração de sólidos suspensos totais (SST), no primeiro reator, em torno de 5000 mg/L, com temperatura controlada (de 25 a 30 graus Celsius) e com tempo de detenção hidráulica (TDH) no primeiro reator de 62 a 16 h e no segundo de 16 a 4 h. O objetivo foi avaliar o desempenho e caracterizar física e microbiologicamente o lodo dos reatores UASB operados em dois estágios. Os resultados obtidos no ensaio 1, com TDH de 62 h no primeiro reator e SST no afluente de 5240 mg/L, mostraram eficiências de remoção de DQO total de 86% e 59% e SST de 82% e 57%, no primeiro e segundo reatores, respectivamente. A eficiência de remoção de DQO total e SST no sistema, no ensaio 1, foi de 95% e 94%. No ensaio 2, com TDH de 31 h no primeiro reator e SST de 5000 mg/L no afluente, observou-se eficiência de remoção de DQO total de 86% e 43% e SST de 85% e 58%, no primeiro e segundo reatores, respectivamente. A eficiência de remoção de DQO total e SST no sistema, no ensaio 2, foi de 92% e 94%. No ensaio 3, com TDH de 16 h no primeiro reator e SST de 5490 mg/L no afluente, observou-se eficiência de remoção de DQO total de 73% e 23% e SST de 65% e 20%, no primeiro e segundo reatores, respectivamente. A eficiência de remoção de DQO total e SST no sistema, no ensaio 3, foi de 79% e 73%. A TCOV aplicada no primeiro reator, no ensaio 1, foi de 4,55 kg DQO total/\'M POT.3\'.d, no ensaio 2 de 8,75 kg DQO total/\'M POT.3\'.d e no ensaio 3 de 18,65 kg DQO total/\'M POT.3\'.d. A produção de \'CH IND.4\' no primeiro reator foi de 17,50 a 68,20 L \'CH IND.4\'/d e no segundo reator de 1,62 a 5,50 L \'CH IND.4\'/d com a diminuição do TDH. Os reatores UASB instalados em série foram eficientes na remoção da fração dissolvida e, principalmente, da fração devido à concentração de SST do afluente. Para TCOV de 4,55 kg DQO/\'M POT.3\'.d no primeiro reator, pôde-se obter eficiências de remoção de DQO total e de SST acima de 90% e de DQO dissolvida acima de 85%. Para TCOV de 18,65 kg DQO/\'M POT.3\'.d no primeiro reator, as eficiências de remoção de DQO total e de SST foram acima de 70% e DQO dissolvida acima de 75%. As maiores produções específicas de metano foram obtidas com TCOV de 2,55 kg DQO/\'M POT.3\'.d para o segundo reator e de 8,65 kg DQO/\'M POT.3\'.d para o primeiro reator. A operação dos reatores UASB com valores de concentração de SST no afluente em torno de 5000 mg/L foram prejudiciais ao processo de granulação do lodo. Os grânulos apresentaram distribuição dispersa das morfologias microbianas ao longo da parede, não caracterizando a divisão em camadas definidas. As arqueas metanogênicas predominantes foram as semelhantes à Methanosaeta. / Two bench scale UASB reactors (volumes of 39,0 and 10,5 L) were operated in sequence, fed with swine wastewater with total suspended solids (TSS) concentration around 5000 mg/L in the first reactor, with controlled temperature (from 25 to 30 Celsius degrees) and operating with hydraulic detention time (HDT) in the first reactor varying from 62 to 16 h and in the second reactor from 16 to 4 h. The objective was to evaluate the performance and to characterize physically and microbiologically the sludge from UASB reactors operated in two stages treating swine wastewater. The results obtained in phase 1, with HDT of 62 h in the first reactor and TSS in the influent of 5240 mg/L, presented total COD removal efficiencies of 86% and 59% and TSS reduction efficiency of 82% and 57%, in the first and second reactors, respectively. The removal efficiency of total COD and TSS in the system, in phase 1, was 95% and 94%, respectively. In the phase 2, using HDT of 31 h in the first reactor and TSS of 5000 mg/L in the influent, it was observed a total COD removal efficiency of 86% and 43% and TSS reduction efficiency of 85% and 58%, in the first and second reactors, respectively. The system removal efficiency of total COD and TSS in phase 2, was 92% and 94%, respectively. In phase 3, with a HDT of 16 h in the first reactor and TSS of 5490 mg/L in the influent, it was observed a total COD removal efficiency of 73% and 23% and TSS reduction efficiency of 65% and 20%, in the first and second reactors, respectively. The total COD removal efficiency and TSS reduction efficiency in the system, in phase 3, was 79% and 73%, respectively. The volumetric organic loading rate (VOLR) applied in the first reactor, in phase 1, was 4,55 kg total COD/\'M POT.3\'.d, in phase 2 was 8,75 kg total COD/\'M POT.3\'.d and in phase 3 was 18,65 kg total COD/\'M POT.3\'.d. The \'CH IND.4\' production in the first reactor was from 17,50 to 68,20 L \'CH IND.4\'/d and in the second reactor from 1,62 to 5,50 L \'CH IND.4\'/d decreasing the HDT in the experiment phases. The UASB reactor installed in sequence were efficient in the dissolved fraction removal and, mainly, to the fraction due to the TSS influent concentration. The total VOL value of 4,55 kg COD/\'M POT.3\'.d was measured in the first reactor, it was achieved TSS and total COD removal efficiencies above 90% and dissolved COD above 85%. For the OVL of 18,65 kg COD/\'M POT.3\'.d in the first reactor the TSS and total COD removal were above 70% and dissolved COD above 75%. The larger methane specific production was obtained with a total VOL of 2,55 kg COD/\'M POT.3\'.d in the second reactor and 8.65 kg COD/\'M POT.3\'.d in the first reactor. The UASB reactors operation with the TSS concentration values of 5000 mg/L in the influent was prejudicing the sludge granulation process. The granules present a microbial morphology disperse distribution that doesn\'t characterize a layers defined distribution. The predominant metonogenic archeas were similar to Methanosaeta.

Joint modeling of longitudinal and survival data has received much attention and is becoming increasingly useful. In clinical studies, longitudinal biomarkers are used to monitor disease progression and to predict survival. These longitudinal measures are often missing at failure times and may be prone to measurement errors. In previous studies these two types of data are frequently analyzed separately where a mixed effects model is used for longitudinal data and a survival model is applied to event outcomes. The argument in favor of a joint model has been the efficient use of the data as the survival information goes into modeling the longitudinal process and vice versa. In this thesis, we present joint maximum likelihood methods, a two stage approach and time dependent covariate methods that link longitudinal data to survival data. First, we use simulation studies to explore and assess the performance of these methods with bias, accuracy and coverage probabilities. Then, we focus on four time dependent methods considering models that are unadjusted and adjusted for time. Finally, we consider mediation analysis for longitudinal and survival data. Mediation analysis is introduced and applied in a research framework based on genetic variants, longitudinal measures and disease risk. We implement accelerated failure time regression using the joint maximum likelihood approach (AFT-joint) and an accelerated failure time regression model using the observed longitudinal measures as time dependent covariates (AFT-observed) to assess the mediated effect. We found that the two stage approach (TSA) performed best at estimating the link parameter. The joint maximum likelihood methods that used the predicted values of the longitudinal measures, similar to the TSA, provided larger estimates. The time dependent covariate methods that used the observed longitudinal measures in the survival analysis underestimated the true estimates. The mediation results showed that the AFT-joint and the AFT-observed underestimated the mediated effect. Comparison of the methods in Framingham Heart Study data revealed similar patterns. We recommend adjusting for time when estimating the association parameter in time dependent Cox and logistic models. Additional work is needed for estimating the mediated effect with longitudinal and survival data.

No description available.

The finite amount of energy carriers affects all of us. It is important to utilize all available sources and to find new sources of energy. The confectionery industry generates both solid and liquid waste during the production of confectioneries, which could be utilized as a substrate for biogas production. However, problems might arise during the biogas process since these kinds of waste are very rich in carbohydrates. The initial breakdown of the substrate would probably cause an accumulation of fermentation products such as volatile fatty acids (VFA) and a low pH. A solution to this might be to use a two-stage process. The first stage would be a pre-fermentation that should be optimized for production of fermentation products such as ethanol and VFA. The aim of this master thesis is to evaluate the biogas potential of confectionery waste products. The confectionery waste products are evaluated using a continuous two-stage process, batch experiments and theoretical calculations of the methane potential. The potential from process wastewater was examined. Depending on COD reduction for a reactor and COD content of process wastewater, an annual amount of 75 000 m3 or of 857 000 m3 of process wastewater is necessary to produce enough biogas for a gas engine to continuously convert the biogas to electricity. A batch experiment evaluating the methane production potential of nine different confectionery waste products from a large confectionery industry gave a range of 430 - 690 NmL/g VS, which is relatively high. A continuous experiment in two lab-scale reactors with a HRT of two days worked satisfactory. The gas production was stable periodically with a carbon dioxide content above 60%. The pH was low (3.4 - 3.6) throughout the experiment for one of the reactors. However, addition of digester sludge from a methane-producing reactor towards the end of the experiment resulted in a higher pH and more VFA available for utilization in the second stage. The main fermentation products were: acetic acid, lactic acid, ethanol and carbon dioxide. A second batch experiment showed that the methane potential was not affected by pre-fermentation. A carbon balance calculation of the process indicates that 57% of the ingoing organic matter is fermented within only two days and ends up in the known fermentation products. The study shows that confectionery waste products are well suited for two-stage anaerobic digestion. / Den ändliga mängden av energibärare påverkar oss alla. Det är därför viktigt att utnyttja alla tillgängliga men även att finna nya energibärare. Konfektyrindustrin generar restprodukter (avfall) både i fast och flytande form, vilka båda kan utnyttjas för produktion av biogas. Det kan dock uppstå problem i biogasprocessen eftersom dessa innehåller en stor mängd kolhydrater. Den initiala nedbrytningen kan ge upphov till en ackumulering av VFA och ett lågt pH. En lösning på detta problem kan vara att använda en två-stegs process. Första steget är en för-fermentering, som ska optimeras för att producera fermentationsprodukter så som etanol och VFA. Syftet med detta arbete är att utvärdera biogaspotentialen från konfektyrrestprodukter genom att använda en kontinuerlig två-stegs process, batchförsök och teoretiska beräkningar av metanpotentialen. Potentialen undersöktes från processvatten. Beroende på COD reduktion i en reaktor och COD innehåll i processvatten, är en årlig mängd av 75 000 m3 eller av 857 000 m3 processvatten nödvändig för att producera en tillräcklig mängd biogas så att en gasmotor kontinuerligt kan omvandla biogasen till elektricitet. Ett batchförsök med nio olika restprodukter från en storskalig konfektyrproducent visade en relativt hög metanpotential (430 - 690 NmL/g VS). Ett kontinuerligt reaktorexperiment genomfördes i laboratorieskala med två reaktorer, där uppehållstiden var två dagar. De två för-fermenteringsreaktorerna presterade tillfredsställande. Gasproduktionen var periodvis stabil med en koldioxidhalt över 60%. pH var lågt (3,4 och 3,6) genom hela experimentet för en av reaktorerna. För den andra reaktorn gjordes tillsatser av reaktormaterial från en metanproducerande reaktor i slutet av experimentet. Dessa tillsatser ökade pH och totalmängden av VFA, som kan utnyttjas i det andra steget. Huvudfermentationsprodukterna är acetat, laktat, etanol och koldioxid. Ytterligare batchförsök visade att för-fermentation inte verkar påverka metanpotentialen för konfektyrrestprodukter. En kolbalans av processen indikerar att 57% av ingående kol återfinns i de kända fermentationsprodukterna inom två dagar. Studiens resultat visar att avfallsprodukter från konfektyrindustrin lämpar sig väl för två-stegs anaerob rötning.

Epidemiological data with disease characteristic information can be modelled in several ways. One way is taking each disease characteristic as a response and constructing binary or polytomous logistic regression model. Second way is using a new response which consists of disease subtypes created by cross-classification of disease characteristic levels, and then constructing polytomous logistic regression model. The former may be disadvantageous since any possible covariation between disease characteristics is neglected, whereas the latter can capture that covariation behaviour. However, cross-classifying the characteristic levels increases the number of categories of response, so that dimensionality problem in parameter space may occur in classical polytomous logistic regression model. A two staged polytomous logistic regression model overcomes that dimensionality problem. In this thesis, study is progressen in two main directions: simulation study and data analysis parts. In simulation study, models that capture the covariation behaviour are compared in terms of the response model parameter estimators. That is, performances of the maximum likelihood estimation (MLE) approach to classical polytomous logistic regression, Bayesian estimation approach to classical polytomous logistic regression and pseudo-conditional likelihood (PCL) estimation approach to two stage polytomous logistic regression are compared in terms of bias and variation of estimators. Results of the simulation study revealed that for small sized sample and small number of disease subtypes, PCL outperforms in terms of bias and variance. For medium scaled size of total disease subtypes situation when sample size is small, PCL performs better than MLE, however when the sample size gets larger MLE has better performance in terms of standard errors of estimates. In addition, sampling variance of PCL estimators of two stage model converges to asymptotic variance faster than the ML estimators of classical polytomous logistic regression model. In data analysis, etiologic heterogeneity in breast cancer subtypes of Turkish female cancer patients is investigated, and the superiority of the two stage polytomous logistic regression model over the classical polytomous logistic model with disease subtypes is represented in terms of the interpretation of parameters and convenience in hypothesis testing.