Markovian Processes, Two-Sided Autoregressions and Finite-Sample Inference for Stationary and Nonstationary Autoregressive Processes.

In this paper, we develop finite-sample inference procedures for stationary and nonstationary autoregressive (AR) models. The method is based on special properties of Markov processes and a split-sample technique. The results on Markovian processes (intercalary independence and truncation) only require the existence of conditional densities. They are proved for possibly nonstationary and/or non-Gaussian multivariate Markov processes. In the context of a linear regression model with AR(1) errors, we show how these results can be used to simplify the distributional properties of the model by conditioning a subset of the data on the remaining observations. This transformation leads to a new model which has the form of a two-sided autoregression to which standard classical linear regression inference techniques can be applied. We show how to derive tests and confidence sets for the mean and/or autoregressive parameters of the model. We also develop a test on the order of an autoregression. We show that a combination of subsample-based inferences can improve the performance of the procedure. An application to U.S. domestic investment data illustrates the method.

Discovery of genes that affect human brain connectivity: A genome-wide analysis of the connectome

Human brain connectivity is disrupted in a wide range of disorders from Alzheimer's disease to autism but little is known about which specific genes affect it. Here we conducted a genome-wide association for connectivity matrices that capture information on the density of fiber connections between 70 brain regions. We scanned a large twin cohort (N=366) with 4-Tesla high angular resolution diffusion imaging (105-gradient HARDI). Using whole brain HARDI tractography, we extracted a relatively sparse 70×70 matrix representing fiber density between all pairs of cortical regions automatically labeled in co-registered anatomical scans. Additive genetic factors accounted for 1-58% of the variance in connectivity between 90 (of 122) tested nodes. We discovered genome-wide significant associations between variants and connectivity. GWAS permutations at various levels of heritability, and split-sample replication, validated our genetic findings. The resulting genes may offer new leads for mechanisms influencing aberrant connectivity and neurodegeneration. © 2012 IEEE.

Daily relative humidity projections in an Indian river basin for IPCC SRES scenarios

A two-stage methodology is developed to obtain future projections of daily relative humidity in a river basin for climate change scenarios. In the first stage, Support Vector Machine (SVM) models are developed to downscale nine sets of predictor variables (large-scale atmospheric variables) for Intergovernmental Panel on Climate Change Special Report on Emissions Scenarios (SRES) (A1B, A2, B1, and COMMIT) to R (H) in a river basin at monthly scale. Uncertainty in the future projections of R (H) is studied for combinations of SRES scenarios, and predictors selected. Subsequently, in the second stage, the monthly sequences of R (H) are disaggregated to daily scale using k-nearest neighbor method. The effectiveness of the developed methodology is demonstrated through application to the catchment of Malaprabha reservoir in India. For downscaling, the probable predictor variables are extracted from the (1) National Centers for Environmental Prediction reanalysis data set for the period 1978-2000 and (2) simulations of the third-generation Canadian Coupled Global Climate Model for the period 1978-2100. The performance of the downscaling and disaggregation models is evaluated by split sample validation. Results show that among the SVM models, the model developed using predictors pertaining to only land location performed better. The R (H) is projected to increase in the future for A1B and A2 scenarios, while no trend is discerned for B1 and COMMIT.

Aplicación de metodología estadística para predecir la mortalidad de pacientes con neumonía

[es]Con esta memoria se pretende crear una herramienta de trabajo que ayude a los médicos a clasificar a los pacientes que padecen neumonía en distintos grupos de riesgo dependiendo de su gravedad. Para ello, hemos estudiado la regresión logística, que relaciona la variable respuesta mortalidad con el resto de variables (datos demográficos, antecedentes médicos y datos clínicos y exploratorios del paciente). Una vez creado el score a partir del modelo de regresión, hemos querido validarlo y compararlo con otros dos que utilizan los neumólogos: el CURB-65 y el FINE.

Pinniped diet composition: a comparison of estimation models

Along the west coast of the United States, the potential impact of increasing pinniped populations on declining salmonid (Oncorhynchus spp.) stocks has become an issue of concern. Fisheries managers need species-specific estimates of consumption by pinnipeds to evaluate their impact on salmonid stocks. To estimate consumption, we developed a model that estimates diet composition by reconstructing prey biomass from fecal samples. We applied the model to data collected from harbor seals (Phoca vitulina) that are present year-round in the lower Columbia River where endangered stocks of salmonids pass as returning adults and as seaward-migrating smolts. Using the same data, we applied the split-sample frequency of occurrence model, which avoids reconstructing biomass by assuming that each fecal sample represents an equal volume of consumption and that within each sample each prey item represents an equal proportion of the volume. The two models for estimating diet composition yielded size-specific differences in consumption estimates that were as large as tenfold for the smallest and largest prey. Conclusions about the impact of harbor seal predation on adult salmonids, some of their largest prey species, remain uncertain without some appropriate rationale or further information (e.g. empirical captive studies) to discriminate between these models.

Testing the stability of the benefit transfer function for discrete choice contingent valuation data

This paper examines the stability of the benefit transfer function across 42 recreational forests in the British Isles. A working definition of reliable function transfer is Put forward, and a suitable statistical test is provided. A novel split sample method is used to test the sensitivity of the models' log-likelihood values to the removal of contingent valuation (CV) responses collected at individual forest sites, We find that a stable function improves Our measure of transfer reliability, but not by much. We conclude that, in empirical Studies on transferability, considerations of function stability are secondary to the availability and quality of site attribute data. Modellers' can study the advantages of transfer function stability vis-a-vis the value of additional information on recreation site attributes. (c) 2008 Elsevier GmbH. All rights reserved.

Identification, Weak Instruments and Statistical Inference in Econometrics

We discuss statistical inference problems associated with identification and testability in econometrics, and we emphasize the common nature of the two issues. After reviewing the relevant statistical notions, we consider in turn inference in nonparametric models and recent developments on weakly identified models (or weak instruments). We point out that many hypotheses, for which test procedures are commonly proposed, are not testable at all, while some frequently used econometric methods are fundamentally inappropriate for the models considered. Such situations lead to ill-defined statistical problems and are often associated with a misguided use of asymptotic distributional results. Concerning nonparametric hypotheses, we discuss three basic problems for which such difficulties occur: (1) testing a mean (or a moment) under (too) weak distributional assumptions; (2) inference under heteroskedasticity of unknown form; (3) inference in dynamic models with an unlimited number of parameters. Concerning weakly identified models, we stress that valid inference should be based on proper pivotal functions —a condition not satisfied by standard Wald-type methods based on standard errors — and we discuss recent developments in this field, mainly from the viewpoint of building valid tests and confidence sets. The techniques discussed include alternative proposed statistics, bounds, projection, split-sampling, conditioning, Monte Carlo tests. The possibility of deriving a finite-sample distributional theory, robustness to the presence of weak instruments, and robustness to the specification of a model for endogenous explanatory variables are stressed as important criteria assessing alternative procedures.

Influência do número de espermatozóides na dose e do refluxo pós-inseminação sobre o desempenho reprodutivo de fêmeas suínas inseminadas intrauterinamente.

O presente trabalho objetivou avaliar o efeito do número de espermatozóides na dose inseminante sobre a taxa de prenhez (TPr) e no número de embriões de fêmeas suínas inseminadas pela técnica intra-uterina (IAU), em uma única inseminação (IA), efetuada no intervalo de até 24 horas antes da ovulação, considerando o refluxo de sêmen pós-inseminação. Foram utilizadas 211 fêmeas pluríparas híbridas (Landrace x Large White), com ordem de parto variando de dois a nove e intervalo desmame-estro de dois a seis dias. O diagnóstico de estro foi realizado duas vezes ao dia. Foram utilizadas doses com volume total de 20 ml contendo 0,25 x 109 (tratamento 1), 0,5 x 109 (tratamento 2) ou 1,0 x 109 (tratamento 3) espermatozóides diluídos em Beltsville Thawing Solution (BTS) produzidas em “split sample”. Quatro machos híbridos (Landrace x Large White x Pietrain) foram utilizados como doadores de sêmen. Foi acompanhada a motilidade (MOT) de uma dose de 100 ml contendo 2,5 x 109 espermatozóides proveniente de cada coleta de sêmen com a finalidade de avaliar a viabilidade espermática durante as 240 horas de armazenamento a 17°C. A coleta do refluxo vulvar, até 60 minutos após a IAU foi realizada com bolsas de colostomia descartáveis fixadas na região peri-vulvar. A determinação do número de espermatozóides contidos no refluxo foi realizada em câmara hemocitométrica. O diagnóstico de gestação foi realizado aos 20-23 dias com auxílio da ultra-sonografia transcutânea em tempo real. As fêmeas prenhes foram abatidas 34-41 dias após a IAU, onde foram coletados o útero e os ovários para contagem dos corpos lúteos e embriões. A passagem do cateter foi possível em todas as fêmeas. Não houve refluxo no momento da realização da IAU. Não houve diferença (P>0,05) no percentual de volume (pvol) e de espermatozóides (psptz) refluídos até 60 minutos após a IAU entre os três tratamentos. O baixo (inferior a 15%) ou alto (superior a 15%) psptz refluídos não influenciou a TPr, mas houve correlação negativa (P=0,0003; R=-0,34) do psptz refluídos com o número de embriões totais (NET). Não houve refluxo em apenas oito fêmeas e a amplitude do pvol variou de 0-136%. No presente trabalho, a MOT das doses inseminantes foi semelhante entre os quatro machos (P>0,05). A TPr não diferiu entre os tratamentos (P=0,36) porém foram verificadas diferenças (P<0,05) entre os machos dentro de cada tratamento. A redução na taxa de prenhez foi mais acentuada para o macho D, quando foram utilizadas doses com menor número de espermatozóides. Não houve efeito do macho ou da interação macho com tratamento sobre o número de embriões (P>0,05). O NET e viáveis diferiu entre o T1 e T2 (P<0,05). Utilizando-se apenas uma única inseminação intra-uterina no intervalo de até 24 horas antes da ovulação com 0,5 x 109 espermatozóides, é possível alcançar índices superiores a 85% de prenhez e 14 embriões, aos 34-41 dias de gestação.

Pretreatment Tables Predicting Pathologic Stage of Locally Advanced Prostate Cancer

BACKGROUND Pretreatment tables for the prediction of pathologic stage have been published and validated for localized prostate cancer (PCa). No such tables are available for locally advanced (cT3a) PCa. OBJECTIVE To construct tables predicting pathologic outcome after radical prostatectomy (RP) for patients with cT3a PCa with the aim to help guide treatment decisions in clinical practice. DESIGN, SETTING, AND PARTICIPANTS This was a multicenter retrospective cohort study including 759 consecutive patients with cT3a PCa treated with RP between 1987 and 2010. INTERVENTION Retropubic RP and pelvic lymphadenectomy. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS Patients were divided into pretreatment prostate-specific antigen (PSA) and biopsy Gleason score (GS) subgroups. These parameters were used to construct tables predicting pathologic outcome and the presence of positive lymph nodes (LNs) after RP for cT3a PCa using ordinal logistic regression. RESULTS AND LIMITATIONS In the model predicting pathologic outcome, the main effects of biopsy GS and pretreatment PSA were significant. A higher GS and/or higher PSA level was associated with a more unfavorable pathologic outcome. The validation procedure, using a repeated split-sample method, showed good predictive ability. Regression analysis also showed an increasing probability of positive LNs with increasing PSA levels and/or higher GS. Limitations of the study are the retrospective design and the long study period. CONCLUSIONS These novel tables predict pathologic stage after RP for patients with cT3a PCa based on pretreatment PSA level and biopsy GS. They can be used to guide decision making in men with locally advanced PCa. PATIENT SUMMARY Our study might provide physicians with a useful tool to predict pathologic stage in locally advanced prostate cancer that might help select patients who may need multimodal treatment.

Psychometric properties of the drinking expectancy questionnaire: A review of the factor structure and a proposed new scoring method

Objective: Expectancies about the outcomes of alcohol consumption are widely accepted as important determinants of drinking. This construct is increasingly recognized as a significant element of psychological interventions for alcohol-related problems. Much effort has been invested in producing reliable and valid instruments to measure this construct for research and clinical purposes, but very few have had their factor structure subjected to adequate validation. Among them, the Drinking Expectancies Questionnaire (DEQ) was developed to address some theoretical and design issues with earlier expectancy scales. Exploratory factor analyses, in addition to validity and reliability analyses, were performed when the original questionnaire was developed. The object of this study was to undertake a confirmatory analysis of the factor structure of the DEQ. Method: Confirmatory factor analysis through LISREL 8 was performed using a randomly split sample of 679 drinkers. Results: Results suggested that a new 5-factor model, which differs slightly from the original 6-factor version, was a more robust measure of expectancies. A new method of scoring the DEQ consistent with this factor structure is presented. Conclusions: The present study shows more robust psychometric properties of the DEQ using the new factor structure.

A tree-based decision model to support prediction of the severity of asthma exacerbations in children

This paper describes the development of a tree-based decision model to predict the severity of pediatric asthma exacerbations in the emergency department (ED) at 2 h following triage. The model was constructed from retrospective patient data abstracted from the ED charts. The original data was preprocessed to eliminate questionable patient records and to normalize values of age-dependent clinical attributes. The model uses attributes routinely collected in the ED and provides predictions even for incomplete observations. Its performance was verified on independent validating data (split-sample validation) where it demonstrated AUC (area under ROC curve) of 0.83, sensitivity of 84%, specificity of 71% and the Brier score of 0.18. The model is intended to supplement an asthma clinical practice guideline, however, it can be also used as a stand-alone decision tool.

An experiment on the parameter uncertainty of hydrological models with different levels of complexity in a climate change context

La possibilité d’estimer l’impact du changement climatique en cours sur le comportement hydrologique des hydro-systèmes est une nécessité pour anticiper les adaptations inévitables et nécessaires que doivent envisager nos sociétés. Dans ce contexte, ce projet doctoral présente une étude sur l’évaluation de la sensibilité des projections hydrologiques futures à : (i) La non-robustesse de l’identification des paramètres des modèles hydrologiques, (ii) l’utilisation de plusieurs jeux de paramètres équifinaux et (iii) l’utilisation de différentes structures de modèles hydrologiques. Pour quantifier l’impact de la première source d’incertitude sur les sorties des modèles, quatre sous-périodes climatiquement contrastées sont tout d’abord identifiées au sein des chroniques observées. Les modèles sont calés sur chacune de ces quatre périodes et les sorties engendrées sont analysées en calage et en validation en suivant les quatre configurations du Different Splitsample Tests (Klemeš, 1986;Wilby, 2005; Seiller et al. (2012);Refsgaard et al. (2014)). Afin d’étudier la seconde source d’incertitude liée à la structure du modèle, l’équifinalité des jeux de paramètres est ensuite prise en compte en considérant pour chaque type de calage les sorties associées à des jeux de paramètres équifinaux. Enfin, pour évaluer la troisième source d’incertitude, cinq modèles hydrologiques de différents niveaux de complexité sont appliqués (GR4J, MORDOR, HSAMI, SWAT et HYDROTEL) sur le bassin versant québécois de la rivière Au Saumon. Les trois sources d’incertitude sont évaluées à la fois dans conditions climatiques observées passées et dans les conditions climatiques futures. Les résultats montrent que, en tenant compte de la méthode d’évaluation suivie dans ce doctorat, l’utilisation de différents niveaux de complexité des modèles hydrologiques est la principale source de variabilité dans les projections de débits dans des conditions climatiques futures. Ceci est suivi par le manque de robustesse de l’identification des paramètres. Les projections hydrologiques générées par un ensemble de jeux de paramètres équifinaux sont proches de celles associées au jeu de paramètres optimal. Par conséquent, plus d’efforts devraient être investis dans l’amélioration de la robustesse des modèles pour les études d’impact sur le changement climatique, notamment en développant les structures des modèles plus appropriés et en proposant des procédures de calage qui augmentent leur robustesse. Ces travaux permettent d’apporter une réponse détaillée sur notre capacité à réaliser un diagnostic des impacts des changements climatiques sur les ressources hydriques du bassin Au Saumon et de proposer une démarche méthodologique originale d’analyse pouvant être directement appliquée ou adaptée à d’autres contextes hydro-climatiques.

Analyse de sensibilité d’un multimodèle hydrologique

Les enjeux hydrologiques modernes, de prévisions ou liés aux changements climatiques, forcent l’exploration de nouvelles approches en modélisation afin de combler les lacunes actuelles et d’améliorer l’évaluation des incertitudes. L’approche abordée dans ce mémoire est celle du multimodèle (MM). L’innovation se trouve dans la construction du multimodèle présenté dans cette étude : plutôt que de caler individuellement des modèles et d’utiliser leur combinaison, un calage collectif est réalisé sur la moyenne des 12 modèles globaux conceptuels sélectionnés. Un des défis soulevés par cette approche novatrice est le grand nombre de paramètres (82) qui complexifie le calage et l’utilisation, en plus d’entraîner des problèmes potentiels d’équifinalité. La solution proposée dans ce mémoire est une analyse de sensibilité qui permettra de fixer les paramètres peu influents et d’ainsi réduire le nombre de paramètres total à caler. Une procédure d’optimisation avec calage et validation permet ensuite d’évaluer les performances du multimodèle et de sa version réduite en plus d’en améliorer la compréhension. L’analyse de sensibilité est réalisée avec la méthode de Morris, qui permet de présenter une version du MM à 51 paramètres (MM51) tout aussi performante que le MM original à 82 paramètres et présentant une diminution des problèmes potentiels d’équifinalité. Les résultats du calage et de la validation avec le « Split-Sample Test » (SST) du MM sont comparés avec les 12 modèles calés individuellement. Il ressort de cette analyse que les modèles individuels, composant le MM, présentent de moins bonnes performances que ceux calés indépendamment. Cette baisse de performances individuelles, nécessaire pour obtenir de bonnes performances globales du MM, s’accompagne d’une hausse de la diversité des sorties des modèles du MM. Cette dernière est particulièrement requise pour les applications hydrologiques nécessitant une évaluation des incertitudes. Tous ces résultats mènent à une amélioration de la compréhension du multimodèle et à son optimisation, ce qui facilite non seulement son calage, mais également son utilisation potentielle en contexte opérationnel.

Simulation studies on the magnetization of (RE)BCO bulk superconductors using various split-coil arrangements

Simulation studies were conducted on the magnetization of (RE)BCO (RE-Ba-Cu-O, where RE represents a rare earth element) bulk superconductors using various split-coil arrangements by solving the critical state equation using the commercial software FlexPDE. A pair of coaxial coils of identical size is identified as an optimum arrangement for practical magnetization at 77K by the zero-field cooling technique. In general, the magnetization process is likely to be most effective when the outer radius of the coils lies between 100% and 50% of the sample radius. A relatively large coil pair is necessary for samples with either a smaller aspect ratio or larger values of J c0. Two different regimes of flux penetration are found to be involved in the magnetization process. For a sufficiently small sample, the penetration field is determined by flux propagation from beneath the coil to the centre of the sample; for a sufficiently large sample, the definitive propagation route is from beneath the coil to the periphery of the sample. Effective split-coil magnetization occurs only in the former regime, and both penetration regimes are completely different from that involved in the solenoidal-coil magnetization process. © 2012 IOP Publishing Ltd.