Sur les tests lisses d'ajustement dans le context des series chronologiques

La plupart des modèles en statistique classique repose sur une hypothèse sur la distribution des données ou sur une distribution sous-jacente aux données. La validité de cette hypothèse permet de faire de l’inférence, de construire des intervalles de confiance ou encore de tester la fiabilité du modèle. La problématique des tests d’ajustement vise à s’assurer de la conformité ou de la cohérence de l’hypothèse avec les données disponibles. Dans la présente thèse, nous proposons des tests d’ajustement à la loi normale dans le cadre des séries chronologiques univariées et vectorielles. Nous nous sommes limités à une classe de séries chronologiques linéaires, à savoir les modèles autorégressifs à moyenne mobile (ARMA ou VARMA dans le cas vectoriel). Dans un premier temps, au cas univarié, nous proposons une généralisation du travail de Ducharme et Lafaye de Micheaux (2004) dans le cas où la moyenne est inconnue et estimée. Nous avons estimé les paramètres par une méthode rarement utilisée dans la littérature et pourtant asymptotiquement efficace. En effet, nous avons rigoureusement montré que l’estimateur proposé par Brockwell et Davis (1991, section 10.8) converge presque sûrement vers la vraie valeur inconnue du paramètre. De plus, nous fournissons une preuve rigoureuse de l’inversibilité de la matrice des variances et des covariances de la statistique de test à partir de certaines propriétés d’algèbre linéaire. Le résultat s’applique aussi au cas où la moyenne est supposée connue et égale à zéro. Enfin, nous proposons une méthode de sélection de la dimension de la famille d’alternatives de type AIC, et nous étudions les propriétés asymptotiques de cette méthode. L’outil proposé ici est basé sur une famille spécifique de polynômes orthogonaux, à savoir les polynômes de Legendre. Dans un second temps, dans le cas vectoriel, nous proposons un test d’ajustement pour les modèles autorégressifs à moyenne mobile avec une paramétrisation structurée. La paramétrisation structurée permet de réduire le nombre élevé de paramètres dans ces modèles ou encore de tenir compte de certaines contraintes particulières. Ce projet inclut le cas standard d’absence de paramétrisation. Le test que nous proposons s’applique à une famille quelconque de fonctions orthogonales. Nous illustrons cela dans le cas particulier des polynômes de Legendre et d’Hermite. Dans le cas particulier des polynômes d’Hermite, nous montrons que le test obtenu est invariant aux transformations affines et qu’il est en fait une généralisation de nombreux tests existants dans la littérature. Ce projet peut être vu comme une généralisation du premier dans trois directions, notamment le passage de l’univarié au multivarié ; le choix d’une famille quelconque de fonctions orthogonales ; et enfin la possibilité de spécifier des relations ou des contraintes dans la formulation VARMA. Nous avons procédé dans chacun des projets à une étude de simulation afin d’évaluer le niveau et la puissance des tests proposés ainsi que de les comparer aux tests existants. De plus des applications aux données réelles sont fournies. Nous avons appliqué les tests à la prévision de la température moyenne annuelle du globe terrestre (univarié), ainsi qu’aux données relatives au marché du travail canadien (bivarié). Ces travaux ont été exposés à plusieurs congrès (voir par exemple Tagne, Duchesne et Lafaye de Micheaux (2013a, 2013b, 2014) pour plus de détails). Un article basé sur le premier projet est également soumis dans une revue avec comité de lecture (Voir Duchesne, Lafaye de Micheaux et Tagne (2016)).

History of the manor and parish of Saleby with Thoresthorpe in the county of Lincoln, with some owners,

Author's autograph copy.

Photo dated 28 November 1973 of Arab auto workers and their supporters in Detroit protest, published in Revolution, January 1974

2 scans - 1of2=image itself, 2of2=reverse with post-it note

Finite mixture regression model with random effects: application to neonatal hospital length of stay

A two-component mixture regression model that allows simultaneously for heterogeneity and dependency among observations is proposed. By specifying random effects explicitly in the linear predictor of the mixture probability and the mixture components, parameter estimation is achieved by maximising the corresponding best linear unbiased prediction type log-likelihood. Approximate residual maximum likelihood estimates are obtained via an EM algorithm in the manner of generalised linear mixed model (GLMM). The method can be extended to a g-component mixture regression model with the component density from the exponential family, leading to the development of the class of finite mixture GLMM. For illustration, the method is applied to analyse neonatal length of stay (LOS). It is shown that identification of pertinent factors that influence hospital LOS can provide important information for health care planning and resource allocation. (C) 2002 Elsevier Science B.V. All rights reserved.

Optimal price regulation in a growth model with monopolistic suppliers of intermediate goods

In this paper we investigate the trade-off faced by regulators who must set a price for an intermediate good somewhere between the marginal cost and the monopoly price. We utilize a growth model with monopolistic suppliers of intermediate goods. Investment in innovation is required to produce a new intermediate good. Marginal cost pricing deters innovation, while monopoly pricing maximizes innovation and economic growth at the cost of some static inefficiency. We demonstrate the existence of a second-best price above the marginal cost but below the monopoly price, which maximizes consumer welfare. Simulation results suggest that substantial reductions in consumption, production, growth, and welfare occur where regulators focus on static efficiency issues by setting prices at or near marginal cost.

Exact solution of the XXZ Gaudin model with generic open boundaries

The XXZ Gaudin model with generic integrable boundaries specified by generic non-diagonal K-matrices is studied. The commuting families of Gaudin operators are diagonalized by the algebraic Bethe ansatz method. The eigenvalues and the corresponding Bethe ansatz equations are obtained. (C) 2004 Elsevier B.V. All rights reserved.

Functional serotonin 5-HT4 receptors in porcine and human ventricular myocardium with increased 5-HT4 mRNA in heart failure

Serotonin (5-hydroxytryptamine, 5-HT) increases contractile force and elicits arrhythmias through 5-HT4 receptors in porcine and human atrium, but its ventricular effects are unknown. We now report functional 5-HT4 receptors in porcine and human ventricle. 5-HT4 mRNA levels were determined in porcine and human ventricles and contractility studied in ventricular trabeculae. Cyclic AMP-dependent protein kinase (PKA) activity was measured in porcine ventricle. Porcine and human ventricles expressed 5-HT4 receptor mRNA. Ventricular 5-HT4(b) mRNA was increased by four times in 20 failing human hearts compared with five donor hearts. 5-HT increased contractile force maximally by 16% (EC50=890 nM) and PKA activity by 20% of the effects of (-)-isoproterenol (200 muM) in ventricular trabeculae from new-born piglets in the presence of the phosphodiesterase-inhibitor 3-isobutyl-1-methylxanthine. In ventricular trabeculae from adult pigs (3-isobutyl-1-methylxanthine present) 5-HT increased force by 32% (EC50=60 nM) and PKA activity by 39% of (-)-iso-proterenol. In right and left ventricular trabeculae from failing hearts, exposed to modified Krebs solution, 5-HT produced variable increases in contractile force in right ventricular trabeculae from 4 out of 6 hearts and in left ventricular trabeculae from 3 out of 3 hearts- range 1-39% of (-)-isoproterenol, average 8%. In 11 left ventricular trabeculae from the failing hearts of four beta-blocker-treated patients, pre-exposed to a relaxant solution with 0.5 mM Ca2+ and 1.2 mM Mg2+ followed by a switch to 2.5 mM Ca2+ and 1 mM Mg2+, 5-HT (1-100 muM, 3-isobutyl-1-melhylxanthine present) consistently increased contractile force and hastened relaxation by 46% and 25% of (-)-isoproterenol respectively. 5-HT caused arrhythmias in three trabeculae from 3 out of I I patients. In the absence of phosphodiesterase inhibitor, 5-HT increased force in two trabeculae, but not in another six trabeculae from 4 patients. All 5-HT responses were blocked by 5-HT4 receptor antagonists. We conclude that phosphodiesterase inhibition uncovers functional ventricular 5-HT4 receptors, coupled to a PKA pathway, through which 5-HT enhances contractility, hastens relaxation and can potentially cause arrhythmias.

Cost-effectiveness of combination peginterferon alpha-2a and ribavirin compared with interferon alpha-2b and ribavirin in patients with chronic hepatitis C

BACKGROUND: Sustained virological response (SVR) is the primary objective in the treatment of chronic hepatitis C (CHC). Results from a recent clinical trial of patients with previously untreated CHC demonstrate that the combination of peginterferon alpha-2a and ribavirin produces a greater SVR than interferon alpha-2b and ribavirin combination therapy. However, the cost-effectiveness of peginterferon alpha-2a plus ribavirin in the U.S. setting has not been investigated. METHODS: A Markov model was developed to investigate cost-effectiveness in patients with CHC using genotype to guide treatment duration. SVR and disease progression parameters were derived from the clinical trials and epidemiologic studies. The impact of treatment on life expectancy and costs were projected for a lifetime. Patients who had an SVR were assumed to remain virus-free for the rest of their lives. In genotype 1 patients, the SVRs were 46% for peginterferon alpha-2a plus ribavirin and 36% for interferon alpha-2b plus ribavirin. In genotype 2/3 patients, the SVRs were 76% for peginterferon alpha-2a plus ribavirin and 61% for interferon alpha-2b plus ribavirin. Quality of life and costs were based on estimates from the literature. All costs were based on published U.S. medical care costs and were adjusted to 2003 U.S. dollars. Costs and benefits beyond the first year were discounted at 3%. RESULTS: In genotype 1, peginterferon alpha-2a plus ribavirin increases quality-adjusted life expectancy (QALY) by 0.70 yr compared to interferon alpha-2b plus ribavirin, producing a cost-effectiveness ratio of $2,600 per QALY gained. In genotype 2/3 patients, peginterferon alpha-2a plus ribavirin increases QALY by 1.05 yr in comparison to interferon alpha-2b plus ribavirin. Peginterferon alpha-2a combination therapy in patients with HCV genotype 2 or 3 is dominant (more effective and cost saving) compared to interferon alpha-2b plus ribavirin. Results weighted by genotype prevalence (75% genotype 1; 25% genotype 2 or 3) also show that peginterferon alpha-2a plus ribavirin is dominant. Peginterferon alpha-2a and ribavirin remained cost-effective (below $16,500 per QALY gained) under sensitivity analyses on key clinical and cost parameters. CONCLUSION: Peginterferon alpha-2a in combination with ribavirin with duration of therapy based on genotype, is cost-effective compared with conventional interferon alpha-2b in combination with ribavirin when given to treatment-naive adults with CHC.

Influence of socioeconomic deprivation on the primary care burden and treatment of patients with a diagnosis of heart failure in general practice in Scotland: population based study

Objectives To examine whether there are socioeconomic gradients in die incidence, prevalence, treatment, and follow up of patients with heart failure in primary care. Design Population based study. Setting 53 general practices (307741 patients) participating in the Scottish continuous morbidity recording project between 1 April 1999 and 31 March 2000. Participants 2186 adults with heart failure. Main outcome measures Comorbid diagnoses, frequency of visits to general practitioner, and prescribed drugs. Results 2186 patients with heart failure were seen (prevalence 7.1 per 1000 population, incidence 2.0 per 1000 population). The age and sex standardised incidence of heart failure increased with greater socioeconomic deprivation, from 1.8 per 1000 population in the most affluent stratum to 2.6 per 1000 population in the most deprived stratum (odds ratio 1.44, P=0.0003). On average, patients were seen 2.4 times yearly, but follow up rates were less frequent with increasing socioeconomic deprivation (from 2.6 yearly in the most affluent subgroup to 2.0 yearly in the most deprived subgroup, P=0.00009). Overall, 812 (80.6%) patients were prescribed diuretics, 396 (39.3%) angiotensin converting enzyme inhibitors, 216 (21.4%) beta blockers, 208 (20.7%) digoxin, and 86 (8.5%) spironolactone. The wide discrepancies in prescribing between different general practices disappeared after adjustment for patient age and sex. Prescribing patterns did not vary by deprivation categories on univariate or multivariate analyses. Conclusions Compared with affluent patients, socioeconomically deprived patients were 44% more likely to develop heart failure but 23% less likely to see their general practitioner on an ongoing basis. Prescribed treatment did not differ across socioeconomic gradients.

The etiology of social phobia: Empirical evidence and an initial model

Research into the etiology of social phobia has lagged far behind that of descriptive and maintaining factors. The current paper reviews data from a variety of sources that have some bearing on questions of the origins of social fears. Areas examined include genetic factors, temperament, childrearing, negative life events, and adverse social experiences. Epidemiological data are examined in detail and factors associated with social phobia such as cognitive distortions and social skills are also covered. The paper concludes with an initial model that draws together some of the current findings and aims to provide a platform for future research directions. (C) 2004 Elsevier Ltd. All rights reserved.

Incidence and risk factors associated with exertional rhabdomyolysis syndrome in National Hunt racehorses in Great Britain

A telephone survey of 51 National Hunt racing yards with 1140 horses in training was made in April and May 2003 to establish the incidence of exertional rhabdomyolysis syndrome during the previous year. A case-control study was used to investigate the risk factors for the syndrome in eight yards selected on the basis that cases had been confirmed by the analysis of serum muscle enzymes. The overall incidence of syndrome was 6 center dot 1 cases per 100 horses per year, and 55 per cent of the yards reported at least one case. The risk factors identified were sex, the average length of the training gallop, and the type of horse (steeplechaser, bumper/unraced or hurdler). There were no significant associations with the horses' temperament, age or Timeform rating.

A carbon activation model with application to longan seed char gasification

In this paper a new structural model is presented to describe the evolution of porosity of char during the gasification process. The model assumes the char structure to be composed of bundles of parallel graphite layers, and the reactivities of each layer with the gasification agent are assumed to be different to represent the different degree of heterogeneity of each layer (i.e. each layer will react with the gasification agent at a different rate). It is this difference in the reactivity that allows micropores to be created during the course of gasification. This simple structural model enables the evolution of pore volume, pore geometrical surface area and the pore size distribution to be described with respect to the extent of char burn-off. The model is tested against the experimental data of gasification of longan seed-derived char with carbon dioxide and it is found that the agreement between the model and the data is reasonably satisfactory, especially the evolution of surface area and pore volume with burn-off.

Vitamin B-12 release from P(HEMA-co-THFMA) in water and SBF: A model drug release study

A model drug release study on the ingress of water and Kokubo simulated body fluid (SBF) into poly(2-hydroxyethyl methacrylate) (THFMA) and its copolymers with tetrahydrofurfuryl methacrylate (THFMA) loaded with vitamin B-12 was undertaken over the temperature range 298-318 K. The polymers were studied as cylinders and were loaded with either 5 or 10 wt-% of the drug. The drug release from the polymers was found to follow a Fickian diffusion mechanism in the early stages of the drug release, with higher normalized release rates at higher temperatures and higher drug loadings. The normalized release rates were also found to be higher for the SBF solution than for water. The copolymer composition was found to have a significant effect on the rate of release of the drug, with the rate falling rapidly between HEMA mole fractions of 1.0 and 0.8, but for lower mole fractions of HEMA the normalized release rate decreased more slowly. This behaviour followed the trend found for the changes in the equilibrium penetrant contents for the copolymers.

Functional-structural plant modelling using a combination of architectural analysis, L-systems and a canonical model of function

This paper presents a new method for producing a functional-structural plant model that simulates response to different growth conditions, yet does not require detailed knowledge of underlying physiology. The example used to present this method is the modelling of the mountain birch tree. This new functional-structural modelling approach is based on linking an L-system representation of the dynamic structure of the plant with a canonical mathematical model of plant function. Growth indicated by the canonical model is allocated to the structural model according to probabilistic growth rules, such as rules for the placement and length of new shoots, which were derived from an analysis of architectural data. The main advantage of the approach is that it is relatively simple compared to the prevalent process-based functional-structural plant models and does not require a detailed understanding of underlying physiological processes, yet it is able to capture important aspects of plant function and adaptability, unlike simple empirical models. This approach, combining canonical modelling, architectural analysis and L-systems, thus fills the important role of providing an intermediate level of abstraction between the two extremes of deeply mechanistic process-based modelling and purely empirical modelling. We also investigated the relative importance of various aspects of this integrated modelling approach by analysing the sensitivity of the standard birch model to a number of variations in its parameters, functions and algorithms. The results show that using light as the sole factor determining the structural location of new growth gives satisfactory results. Including the influence of additional regulating factors made little difference to global characteristics of the emergent architecture. Changing the form of the probability functions and using alternative methods for choosing the sites of new growth also had little effect. (c) 2004 Elsevier B.V. All rights reserved.

Evaluation of eight spring onion genotypes, sulphur nutrition and soil-type effects with an electronic nose

Genotype, sulphur (S) nutrition and soil-type effects on spring onion quality were assessed using a 32-conducting polymer sensor E-nose. Relative changes in sensor resistance ratio (% dR/R) varied among eight spring onion genotypes. The % dR/R was reduced by S application in four of the eight genotypes. For the other four genotypes, S application gave no change in % dR/R in three, and increased % dR/R in the other. E-nose classification of headspace volatiles by a two-dimensional principal component analysis (PCA) plot for spring onion genotypes differed for S fertilisation vs. no S fertilisation. Headspace volatiles data set clusters for cv. 'White Lisbon' grown on clay or on sandy loam overlapped when 2.9 [Mahalanobis distance value (D2) = 1.6], or 5.8-(D2 = 0.3) kg S ha-1 was added. In contrast, clear separation (D2 = 7.5) was recorded for headspace volatile clusters for 0 kg S hd-1 on clay vs. sandy loam. Addition of 5.8 kg S ha-1 increased pyruvic acid content (mmole g-1 fresh weight) by 1.7-fold on average across the eight genotypes. However, increased S from 2.9 to 5.8 kg ha-1 did not significantly (P > 0.05) influence % dR/R, % dry matter (DM) or total soluble solids (TSS) contents, but significantly (P < 0.05) increased pyruvic acid content. TSS was significantly (P < 0.05) reduced by S addition, while % DM was unaffected. In conclusion, the 32-conducting polymer E-nose discerned differences in spring onion quality that were attributable to genotype and to variations in growing conditions as shown by the significant (P < 0.05) interaction effects for % dR/R.