Model specification tests in nonparametric stochastic regression models

In this paper, we consider testing for additivity in a class of nonparametric stochastic regression models. Two test statistics are constructed and their asymptotic distributions are established. We also conduct a small sample study for one of the test statistics through a simulated example. (C) 2002 Elsevier Science (USA).

Indirect evidence of density-dependent population regulation in Aponomma hydrosauri (Acari : Ixodidae), an ectoparasite of reptiles

The extent to which density-dependent processes regulate natural populations is the subject of an ongoing debate. We contribute evidence to this debate showing that density-dependent processes influence the population dynamics of the ectoparasite Aponomma hydrosauri (Acari: Ixodidae), a tick species that infests reptiles in Australia. The first piece of evidence comes from an unusually long-term dataset on the distribution of ticks among individual hosts. If density-dependent processes are influencing either host mortality or vital rates of the parasite population, and those distributions can be approximated with negative binomial distributions, then general host-parasite models predict that the aggregation coefficient of the parasite distribution will increase with the average intensity of infections. We fit negative binomial distributions to the frequency distributions of ticks on hosts, and find that the estimated aggregation coefficient k increases with increasing average tick density. This pattern indirectly implies that one or more vital rates of the tick population must be changing with increasing tick density, because mortality rates of the tick's main host, the sleepy lizard, Tiliqua rugosa, are unaffected by changes in tick burdens. Our second piece of evidence is a re-analysis of experimental data on the attachment success of individual ticks to lizard hosts using generalized linear modelling. The probability of successful engorgement decreases with increasing numbers of ticks attached to a host. This is direct evidence of a density-dependent process that could lead to an increase in the aggregation coefficient of tick distributions described earlier. The population-scale increase in the aggregation coefficient is indirect evidence of a density-dependent process or processes sufficiently strong to produce a population-wide pattern, and thus also likely to influence population regulation. The direct observation of a density-dependent process is evidence of at least part of the responsible mechanism.

Dynamical behaviour on the parameter space: new populational growth models proportional to beta densities

We present new populational growth models, generalized logistic models which are proportional to beta densities with shape parameters p and 2, where p > 1, with Malthusian parameter r. The complex dynamical behaviour of these models is investigated in the parameter space (r, p), in terms of topological entropy, using explicit methods, when the Malthusian parameter r increases. This parameter space is split into different regions, according to the chaotic behaviour of the models.

Prognostic modelling of breast cancer patients: a benchmark of predictive models with external validation

Dissertação apresentada para obtenção do Grau de Doutor em Engenharia Electrotécnica e de Computadores – Sistemas Digitais e Percepcionais pela Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia

Osteoporotic hip fractures: Bisphosphonates sales and observed turning point in trend. A population-based retrospective study

The aim is to examine the temporal trends of hip fracture incidence in Portugal by sex and age groups, and explore the relation with anti-osteoporotic medication. From the National Hospital Discharge Database, we selected from 1st January 2000 to 31st December 2008, 77,083 hospital admissions (77.4% women) caused by osteoporotic hip fractures (low energy, patients over 49 years-age), with diagnosis codes 820.x of ICD 9-CM. The 2001 Portuguese population was used as standard to calculate direct age-standardized incidence rates (ASIR) (100,000 inhabitants). Generalized additive and linear models were used to evaluate and quantify temporal trends of age specific rates (AR), by sex. We identified 2003 as a turning point in the trend of ASIR of hip fractures in women. After 2003, the ASIR in women decreased on average by 10.3 cases/100,000 inhabitants, 95% CI (− 15.7 to − 4.8), per 100,000 anti-osteoporotic medication packages sold. For women aged 65–69 and 75–79 we identified the same turning point. However, for women aged over 80, the year 2004 marked a change in the trend, from an increase to a decrease. Among the population aged 70–74 a linear decrease of incidence rate (95% CI) was observed in both sexes, higher for women: − 28.0% (− 36.2 to − 19.5) change vs − 18.8%, (− 32.6 to − 2.3). The abrupt turning point in the trend of ASIR of hip fractures in women is compatible with an intervention, such as a medication. The trends were different according to gender and age group, but compatible with the pattern of bisphosphonates sales.

Comparing the performance of geostatistical models with additional information from covariates for sewage plume characterization

In this work, kriging with covariates is used to model and map the spatial distribution of salinity measurements gathered by an autonomous underwater vehicle in a sea outfall monitoring campaign aiming to distinguish the effluent plume from the receiving waters and characterize its spatial variability in the vicinity of the discharge. Four different geostatistical linear models for salinity were assumed, where the distance to diffuser, the west-east positioning, and the south-north positioning were used as covariates. Sample variograms were fitted by the Mat`ern models using weighted least squares and maximum likelihood estimation methods as a way to detect eventual discrepancies. Typically, the maximum likelihood method estimated very low ranges which have limited the kriging process. So, at least for these data sets, weighted least squares showed to be the most appropriate estimation method for variogram fitting. The kriged maps show clearly the spatial variation of salinity, and it is possible to identify the effluent plume in the area studied. The results obtained show some guidelines for sewage monitoring if a geostatistical analysis of the data is in mind. It is important to treat properly the existence of anomalous values and to adopt a sampling strategy that includes transects parallel and perpendicular to the effluent dispersion.

Avaliação do impacto na circulação e risco de acidente em estradas em obras

Dissertação de mestrado integrado em Engenharia Civil

A priori ratemaking using bivariate poisson regression models

In automobile insurance, it is useful to achieve a priori ratemaking by resorting to gene- ralized linear models, and here the Poisson regression model constitutes the most widely accepted basis. However, insurance companies distinguish between claims with or without bodily injuries, or claims with full or partial liability of the insured driver. This paper exa- mines an a priori ratemaking procedure when including two di®erent types of claim. When assuming independence between claim types, the premium can be obtained by summing the premiums for each type of guarantee and is dependent on the rating factors chosen. If the independence assumption is relaxed, then it is unclear as to how the tari® system might be a®ected. In order to answer this question, bivariate Poisson regression models, suitable for paired count data exhibiting correlation, are introduced. It is shown that the usual independence assumption is unrealistic here. These models are applied to an automobile insurance claims database containing 80,994 contracts belonging to a Spanish insurance company. Finally, the consequences for pure and loaded premiums when the independence assumption is relaxed by using a bivariate Poisson regression model are analysed.

Evaluating the ability of habitat suitability models to predict species presences.

Models predicting species spatial distribution are increasingly applied to wildlife management issues, emphasising the need for reliable methods to evaluate the accuracy of their predictions. As many available datasets (e.g. museums, herbariums, atlas) do not provide reliable information about species absences, several presence-only based analyses have been developed. However, methods to evaluate the accuracy of their predictions are few and have never been validated. The aim of this paper is to compare existing and new presenceonly evaluators to usual presence/absence measures. We use a reliable, diverse, presence/absence dataset of 114 plant species to test how common presence/absence indices (Kappa, MaxKappa, AUC, adjusted D-2) compare to presenceonly measures (AVI, CVI, Boyce index) for evaluating generalised linear models (GLM). Moreover we propose a new, threshold-independent evaluator, which we call "continuous Boyce index". All indices were implemented in the B10MAPPER software. We show that the presence-only evaluators are fairly correlated (p > 0.7) to the presence/absence ones. The Boyce indices are closer to AUC than to MaxKappa and are fairly insensitive to species prevalence. In addition, the Boyce indices provide predicted-toexpected ratio curves that offer further insights into the model quality: robustness, habitat suitability resolution and deviation from randomness. This information helps reclassifying predicted maps into meaningful habitat suitability classes. The continuous Boyce index is thus both a complement to usual evaluation of presence/absence models and a reliable measure of presence-only based predictions.

Air pollution exposure during pregnancy and reduced birth size: a prospective birth cohort study in Valencia, Spain

Background Maternal exposure to air pollution has been related to fetal growth in a number of recent scientific studies. The objective of this study was to assess the association between exposure to air pollution during pregnancy and anthropometric measures at birth in a cohort in Valencia, Spain. Methods Seven hundred and eighty-five pregnant women and their singleton newborns participated in the study. Exposure to ambient nitrogen dioxide (NO2) was estimated by means of land use regression. NO2 spatial estimations were adjusted to correspond to relevant pregnancy periods (whole pregnancy and trimesters) for each woman. Outcome variables were birth weight, length, and head circumference (HC), along with being small for gestational age (SGA). The association between exposure to residential outdoor NO2 and outcomes was assessed controlling for potential confounders and examining the shape of the relationship using generalized additive models (GAM). Results For continuous anthropometric measures, GAM indicated a change in slope at NO2 concentrations of around 40 μg/m3. NO2 exposure >40 μg/m3 during the first trimester was associated with a change in birth length of -0.27 cm (95% CI: -0.51 to -0.03) and with a change in birth weight of -40.3 grams (-96.3 to 15.6); the same exposure throughout the whole pregnancy was associated with a change in birth HC of -0.17 cm (-0.34 to -0.003). The shape of the relation was seen to be roughly linear for the risk of being SGA. A 10 μg/m3 increase in NO2 during the second trimester was associated with being SGA-weight, odds ratio (OR): 1.37 (1.01-1.85). For SGA-length the estimate for the same comparison was OR: 1.42 (0.89-2.25). Conclusions Prenatal exposure to traffic-related air pollution may reduce fetal growth. Findings from this study provide further evidence of the need for developing strategies to reduce air pollution in order to prevent risks to fetal health and development.

Time-series regression models to study the short-term effects of environmental factors on health

Time series regression models are especially suitable in epidemiology for evaluating short-term effects of time-varying exposures on health. The problem is that potential for confounding in time series regression is very high. Thus, it is important that trend and seasonality are properly accounted for. Our paper reviews the statistical models commonly used in time-series regression methods, specially allowing for serial correlation, make them potentially useful for selected epidemiological purposes. In particular, we discuss the use of time-series regression for counts using a wide range Generalised Linear Models as well as Generalised Additive Models. In addition, recently critical points in using statistical software for GAM were stressed, and reanalyses of time series data on air pollution and health were performed in order to update already published. Applications are offered through an example on the relationship between asthma emergency admissions and photochemical air pollutants

Correcting Measurement Error Bias in Interaction Models with Small Samples

Several methods have been suggested to estimate non-linear models with interaction terms in the presence of measurement error. Structural equation models eliminate measurement error bias, but require large samples. Ordinary least squares regression on summated scales, regression on factor scores and partial least squares are appropriate for small samples but do not correct measurement error bias. Two stage least squares regression does correct measurement error bias but the results strongly depend on the instrumental variable choice. This article discusses the old disattenuated regression method as an alternative for correcting measurement error in small samples. The method is extended to the case of interaction terms and is illustrated on a model that examines the interaction effect of innovation and style of use of budgets on business performance. Alternative reliability estimates that can be used to disattenuate the estimates are discussed. A comparison is made with the alternative methods. Methods that do not correct for measurement error bias perform very similarly and considerably worse than disattenuated regression

Linear response functions for a vibrational configuration interaction state

Linear response functions are implemented for a vibrational configuration interaction state allowing accurate analytical calculations of pure vibrational contributions to dynamical polarizabilities. Sample calculations are presented for the pure vibrational contributions to the polarizabilities of water and formaldehyde. We discuss the convergence of the results with respect to various details of the vibrational wave function description as well as the potential and property surfaces. We also analyze the frequency dependence of the linear response function and the effect of accounting phenomenologically for the finite lifetime of the excited vibrational states. Finally, we compare the analytical response approach to a sum-over-states approach

Variational calculation of vibrational linear and nonlinear optical properties

A variational approach for reliably calculating vibrational linear and nonlinear optical properties of molecules with large electrical and/or mechanical anharmonicity is introduced. This approach utilizes a self-consistent solution of the vibrational Schrödinger equation for the complete field-dependent potential-energy surface and, then, adds higher-level vibrational correlation corrections as desired. An initial application is made to static properties for three molecules of widely varying anharmonicity using the lowest-level vibrational correlation treatment (i.e., vibrational Møller-Plesset perturbation theory). Our results indicate when the conventional Bishop-Kirtman perturbation method can be expected to break down and when high-level vibrational correlation methods are likely to be required. Future improvements and extensions are discussed

Neocortical maturation during adolescence: change in neuronal soma dimension.

During adolescence, cognitive abilities increase robustly. To search for possible related structural alterations of the cerebral cortex, we measured neuronal soma dimension (NSD = width times height), cortical thickness and neuronal densities in different types of neocortex in post-mortem brains of five 12-16 and five 17-24 year-olds (each 2F, 3M). Using a generalized mixed model analysis, mean normalized NSD comparing the age groups shows layer-specific change for layer 2 (p < .0001) and age-related differences between categorized type of cortex: primary/primary association cortex (BA 1, 3, 4, and 44) shows a generalized increase; higher-order regions (BA 9, 21, 39, and 45) also show increase in layers 2 and 5 but decrease in layers 3, 4, and 6 while limbic/orbital cortex (BA 23, 24, and 47) undergoes minor decrease (BA 1, 3, 4, and 44 vs. BA 9, 21, 39, and 45: p = .036 and BA 1, 3, 4, and 44 vs. BA 23, 24, and 47: p = .004). These data imply the operation of cortical layer- and type-specific processes of growth and regression adding new evidence that the human brain matures during adolescence not only functionally but also structurally.