Les modèles vectoriels et multiplicatifs avec erreurs non-négatives de séries chronologiques.

L'objectif du présent mémoire vise à présenter des modèles de séries chronologiques multivariés impliquant des vecteurs aléatoires dont chaque composante est non-négative. Nous considérons les modèles vMEM (modèles vectoriels et multiplicatifs avec erreurs non-négatives) présentés par Cipollini, Engle et Gallo (2006) et Cipollini et Gallo (2010). Ces modèles représentent une généralisation au cas multivarié des modèles MEM introduits par Engle (2002). Ces modèles trouvent notamment des applications avec les séries chronologiques financières. Les modèles vMEM permettent de modéliser des séries chronologiques impliquant des volumes d'actif, des durées, des variances conditionnelles, pour ne citer que ces applications. Il est également possible de faire une modélisation conjointe et d'étudier les dynamiques présentes entre les séries chronologiques formant le système étudié. Afin de modéliser des séries chronologiques multivariées à composantes non-négatives, plusieurs spécifications du terme d'erreur vectoriel ont été proposées dans la littérature. Une première approche consiste à considérer l'utilisation de vecteurs aléatoires dont la distribution du terme d'erreur est telle que chaque composante est non-négative. Cependant, trouver une distribution multivariée suffisamment souple définie sur le support positif est plutôt difficile, au moins avec les applications citées précédemment. Comme indiqué par Cipollini, Engle et Gallo (2006), un candidat possible est une distribution gamma multivariée, qui impose cependant des restrictions sévères sur les corrélations contemporaines entre les variables. Compte tenu que les possibilités sont limitées, une approche possible est d'utiliser la théorie des copules. Ainsi, selon cette approche, des distributions marginales (ou marges) peuvent être spécifiées, dont les distributions en cause ont des supports non-négatifs, et une fonction de copule permet de tenir compte de la dépendance entre les composantes. Une technique d'estimation possible est la méthode du maximum de vraisemblance. Une approche alternative est la méthode des moments généralisés (GMM). Cette dernière méthode présente l'avantage d'être semi-paramétrique dans le sens que contrairement à l'approche imposant une loi multivariée, il n'est pas nécessaire de spécifier une distribution multivariée pour le terme d'erreur. De manière générale, l'estimation des modèles vMEM est compliquée. Les algorithmes existants doivent tenir compte du grand nombre de paramètres et de la nature élaborée de la fonction de vraisemblance. Dans le cas de l'estimation par la méthode GMM, le système à résoudre nécessite également l'utilisation de solveurs pour systèmes non-linéaires. Dans ce mémoire, beaucoup d'énergies ont été consacrées à l'élaboration de code informatique (dans le langage R) pour estimer les différents paramètres du modèle. Dans le premier chapitre, nous définissons les processus stationnaires, les processus autorégressifs, les processus autorégressifs conditionnellement hétéroscédastiques (ARCH) et les processus ARCH généralisés (GARCH). Nous présentons aussi les modèles de durées ACD et les modèles MEM. Dans le deuxième chapitre, nous présentons la théorie des copules nécessaire pour notre travail, dans le cadre des modèles vectoriels et multiplicatifs avec erreurs non-négatives vMEM. Nous discutons également des méthodes possibles d'estimation. Dans le troisième chapitre, nous discutons les résultats des simulations pour plusieurs méthodes d'estimation. Dans le dernier chapitre, des applications sur des séries financières sont présentées. Le code R est fourni dans une annexe. Une conclusion complète ce mémoire.

Interacciones de género estudiante-profesor, deserción y rendimiento académico en Colombia

En este trabajo se estima el efecto del género del profesor sobre la deserción y el rendimiento educativo de los estudiantes en Colombia durante el periodo 2009-2012. La estrategia empírica se fundamenta en un modelo de regresión lineal que establece la relación entre la proporción de profesoras interactuada con el género del estudiante. Los resultados sugieren que existe un sesgo de selección debido a que las profesoras aumentan la probabilidad de que las niñas finalicen la educación media, lo que implica que la composición de habilidades entre hombres y mujeres no es la misma. Luego de corregir este sesgo de selección, se encuentra un resultado significativo en el género del profesor. Un aumento de una desviación estándar en la proporción de profesoras incrementa en 0.01 desviaciones estándar el puntaje de los niños en la prueba de matemáticas y el puntaje de las niñas en la prueba de lenguaje. Este trabajo utiliza información proveniente de la prueba Saber 11, la Resolución 166 y el concurso docente.

Taking the LEED? Analyzing spatial variations in market penetration rates of eco-labeled properties

This paper investigates the impact of policies to promote the adoption of LEED-certified buildings across CBSA in the United States. Drawing upon a unique database that combines data from a large number of sources and using a number of regression procedures, the determinants of the proportion LEED-certified space for more than 170 CBSA in the US is modeled. LEED-certified space still accounts for a relatively small proportion of commercial stock in all markets. The average proportion is less than 1%. There is no conclusive evidence of a positive impact of policy intervention on the levels of LEED-certified space. However, after accounting for bias introduced by non-random assignment of policies, we find preliminary evidence of a positive impact of city-level green building incentives. There is a significant positive association between market size and indicators of economic vitality on proportions of LEED-certified space.

Variance component estimates applying random regression models for test-day milk yield in Caracu heifers (Bos taurus Artiodactyla, Bovidae)

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Ximelagatran versus warfarin for prophylaxis of venous thromboembolism in major orthopedic surgery: systematic review of randomized controlled trials

INTRODUÇÃO: O ximelagatrano foi recentemente estudada para profilaxia do tromboembolismo venoso (TEV). OBJETIVO: Avaliar se o ximelagatrano comparado com a varfarina melhora a profilaxia do TEV em pacientes submetidos à cirurgia ortopédica do joelho. FONTE DE DADOS: Estudos randomizados identificados por pesquisa eletrônica na literatura médica, até 2006, cujos dados foram compilados no programa Review Manager, versão 4.2.5. RESULTADOS: Foram incluídos três estudos randomizados bem conduzidos envolvendo 4.914 participantes. Foram definidos dois sub-grupos com dosagens diferentes de ximelagatrano (24 mg and 36 mg, duas vezes ao dia). O tratamento com ximelagatrano mostrou freqüência significantemente menor de TEV que o tratamento com varfarina, mas somente na dosagem de 36-mg [risco relativo, RR 0.72 ([intervalo de confiança, IC, 95% 0.64, 0.81), p < 0.00001]. A freqüência de TEV no sub-grupo de 24-mg foi similar a da varfarina [RR 0.86 (IC 95% 0.73, 1.01), p = 0.06]. Para TEV maior, embolia pulmonar, sangramento e sangramento maior não houve diferença entre varfarina e a ximelagatrano. Ao final do tratamento, a elevação da alanino-aminotransferase (ALT) foi menos freqüente no sub-grupo de 24 mg de ximelagatrano que no grupo da varfarina [RR 0.33 (IC 95% 0.12, 0.91) p = 0.03], mas no período de acompanhamento essa elevação foi maior com 36 mg de ximelagatrano [RR 6.97 (IC 95% 1.26, 38.50) p = 0.03]. CONCLUSÃO: O ximelagatrano foi mais efetivo que a varfarina quando usado em dosagens maiores (36 mg, 2 vezes ao dia), mas às expensas de aumento de enzimas hepáticas no período de acompanhamento.

The effect of refuge on Dermestes ater (Coleoptera : Dermestidae) predation on Musca domestica (Diptera : Muscidae): Refuge for prey or the predator?

It is believed that habitat heterogeneity can change the extent of predator-prey interactions. Therefore, in this study we examined the effect of habitat heterogeneity (characterized here as an addition of refuge) on D. ater predation on M. domestica. Predation of D. ater on M. domestica larvae was carried out in experimental habitats with and without refuge, and examined at different prey densities. The number of prey eaten by beetles over 24 h of predator-prey interaction was recorded, and we investigated the strength of interaction between prey and predator in both experimental habitats by determining predator functional response. The mean number of prey eaten by beetles in the presence of refuge was significantly higher than in the absence of refuge. Females had greater weight gains than males. Logistic regression analyses revealed the type II functional response for both experimental habitats, even though data did not fit well into the random predator model. Results suggest that the addition of refuge in fact enhanced predation, as prey consumption increased in the presence of refuge. Predators kept in the presence of refuge also consumed more prey at high prey densities. Thus, we concluded that the addition of refuge was an important component mediating D. ater-M. domestica population interactions. Refuge actually acted as a refuge for predators from prey, since prey behaviors detrimental to predators were reduced in this case.

Two-Way Selection for Daily Gain and Feed Conversion in a Composite Rabbit Population

We conducted a two-way selection experiment in a composite rabbit population to investigate the responses to selection for postweaning ADG and feed conversion (FC). Two generations of crossing, followed by four generations of random pair matings, preceded three generations of selection. Selection was practiced within four lines: high-feed conversion (HFC), low-feed conversion (LFC), high gain (HG), and low gain (LG). Data on 1,446 rabbits from the random mating and selection generations were fitted to an animal model to estimate heritabilities of and the genetic correlation between ADG and FC. The two-trait model included rabbit and common litter random effects and line, generation, and sex fixed effects. Estimates of heritability of ADG and FC were .48 and .29, respectively, and the genetic correlation between them was -.82. Common litter environmental effects accounted for a proportion of .11 and . 13 of the phenotypic variation of the two traits, respectively. For ADG (in g/d) the regressions of mean breeding values on generation number during the selection period were 1.23 ± .12 (P < .01) in the HG line and -.86 ± .12 (P < .01) in the LG line; the regressions for FC (in g feed/g gain) were -.07 ± .01 (P < .01) in the HFC line and .03 ± .01 (P < .05) in the LFC line. Selection for ADG was effective in improving ADG and FC.

Robust linear mixed models with normal/independent distributions and Bayesian MCMC implementation

Linear mixed effects models have been widely used in analysis of data where responses are clustered around some random effects, so it is not reasonable to assume independence between observations in the same cluster. In most biological applications, it is assumed that the distributions of the random effects and of the residuals are Gaussian. This makes inferences vulnerable to the presence of outliers. Here, linear mixed effects models with normal/independent residual distributions for robust inferences are described. Specific distributions examined include univariate and multivariate versions of the Student-t, the slash and the contaminated normal. A Bayesian framework is adopted and Markov chain Monte Carlo is used to carry out the posterior analysis. The procedures are illustrated using birth weight data on rats in a texicological experiment. Results from the Gaussian and robust models are contrasted, and it is shown how the implementation can be used for outlier detection. The thick-tailed distributions provide an appealing robust alternative to the Gaussian process in linear mixed models, and they are easily implemented using data augmentation and MCMC techniques.

Interferometria óptica aplicada à medição de amplitudes de vibração nanométricas em piezoatuadores flextensionais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Parâmetros genéticos para produção de leite e persistência de lactações múltiplas na raça Gir

Pós-graduação em Genética e Melhoramento Animal - FCAV

Distribuição espacial de Phytophthora nicotianae da cultura da cebola em amostras de solo e reação de cultivares ao patógeno

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Network-based stochastic semisupervised learning

Semisupervised learning is a machine learning approach that is able to employ both labeled and unlabeled samples in the training process. In this paper, we propose a semisupervised data classification model based on a combined random-preferential walk of particles in a network (graph) constructed from the input dataset. The particles of the same class cooperate among themselves, while the particles of different classes compete with each other to propagate class labels to the whole network. A rigorous model definition is provided via a nonlinear stochastic dynamical system and a mathematical analysis of its behavior is carried out. A numerical validation presented in this paper confirms the theoretical predictions. An interesting feature brought by the competitive-cooperative mechanism is that the proposed model can achieve good classification rates while exhibiting low computational complexity order in comparison to other network-based semisupervised algorithms. Computer simulations conducted on synthetic and real-world datasets reveal the effectiveness of the model.

Studying the influence of a secondary variable in Collocated Cokriging estimates

In this paper the influence of a secondary variable as a function of the correlation with the primary variable for collocated cokriging is examined. For this study five exhaustive data sets were generated in computer, from which samples with 60 and 104 data points were drawn using the stratified random sampling method. These exhaustive data sets were generated departing from a pair of primary and secondary variables showing a good correlation. Then successive sets were generated by adding an amount of white noise in such a way that the correlation gets poorer. Using these samples, it was possible to find out how primary and secondary information is used to estimate an unsampled location according to the correlation level.

Sacral neuromodulation for neurogenic lower urinary tract dysfunction: systematic review and meta-analysis

Context Treatment of neurogenic lower urinary tract dysfunction (LUTD) is a challenge, because conventional therapies often fail. Sacral neuromodulation (SNM) has become a well-established therapy for refractory non-neurogenic LUTD, but its value in patients with a neurologic cause is unclear. Objective To assess the efficacy and safety of SNM for neurogenic LUTD. Evidence acquisition Studies were identified by electronic search of PubMed, EMBASE, and ScienceDirect (on 15 April 2010) and hand search of reference lists and review articles. SNM articles were included if they reported on efficacy and/or safety of tested and/or permanently implanted patients suffering from neurogenic LUTD. Two reviewers independently selected studies and extracted data. Study estimates were pooled using Bayesian random-effects meta-analysis. Evidence synthesis Of the 26 independent studies (357 patients) included, the evidence level ranged from 2b to 4 according to the Oxford Centre for Evidence-Based Medicine. Half (n = 13) of the included studies reported data on both test phase and permanent SNM; the remaining studies were confined to test phase (n = 4) or permanent SNM (n = 9). The pooled success rate was 68% for the test phase (95% credibility interval [CrI], 50–87) and 92% (95% CrI, 81–98%) for permanent SNM, with a mean follow-up of 26 mo. The pooled adverse event rate was 0% (95% CrI, 0–2%) for the test phase and 24% (95% CrI, 6–48%) for permanent SNM. Conclusions There is evidence indicating that SNM may be effective and safe for the treatment of patients with neurogenic LUTD. However, the number of investigated patients is low with high between-study heterogeneity, and there is a lack of randomised, controlled trials. Thus, well-designed, adequately powered studies are urgently needed before more widespread use of SNM for neurogenic LUTD can be recommended.

Fasting plasma glucose predicts survival and angiographic progression in high-risk postmenopausal women with coronary artery disease

BACKGROUND: We studied the association of baseline fasting plasma glucose (FPG) levels with survival and coronary artery disease (CAD) progression among postmenopausal women without unstable angina. METHODS: Women were recruited from seven centers in the Women's Angiographic Vitamin and Estrogen Trial (WAVE) (n = 423). Event follow-up was available for 400 women (65.1 +/- 8.5 years, 66% white, 92% hypertensive, 19% smokers, 67% hypercholesterolemic). Thirty-eight percent of the women had diabetes or FPG > 125 mg/dL, and 21% had a fasting glucose 100-125 mg/dL. Follow-up angiography was performed in 304 women. Cox regression was used to model survival from a composite outcome of death or myocardial infarction (D/MI, 26 events; median follow-up 2.4 years). Angiographic progression was analyzed quantitatively using linear regression accounting for baseline minimum lumen diameter (MLD), follow-up time, and intrasubject correlations using generalized estimating equations. Regression analyses were adjusted for follow-up time, baseline age, treatment assignment, and Framingham risk (excluding diabetes). RESULTS: Women with impaired fasting glucose/diabetes mellitus (IFG/DM) had a relative risk (RR) of D/MI of 4.2 ( p = 0.009). In all women, each 10 mg/dL increase in FPG was associated with an 11% increase ( p < 0.001) in the hazard of D/MI. Each 10 mg/dL increase in FPG was associated with a 6.8 mum decrease in MLD over the follow-up period ( p = 0.005). CONCLUSIONS: Higher FPG is associated with increased risk of D/MI and greater narrowing of the coronary lumen in women with CAD. Aggressive monitoring of glucose levels may be beneficial for secondary CAD prevention.