Forecasting time series combining Holt-Winters and bootstrap approaches

Publicado em "AIP Conference Proceedings", Vol. 1648

Specification and testing of multiplicative time-varying GARCH models with applications

In this article, we develop a specification technique for building multiplicative time-varying GARCH models of Amado and Teräsvirta (2008, 2013). The variance is decomposed into an unconditional and a conditional component such that the unconditional variance component is allowed to evolve smoothly over time. This nonstationary component is defined as a linear combination of logistic transition functions with time as the transition variable. The appropriate number of transition functions is determined by a sequence of specification tests. For that purpose, a coherent modelling strategy based on statistical inference is presented. It is heavily dependent on Lagrange multiplier type misspecification tests. The tests are easily implemented as they are entirely based on auxiliary regressions. Finite-sample properties of the strategy and tests are examined by simulation. The modelling strategy is illustrated in practice with two real examples: an empirical application to daily exchange rate returns and another one to daily coffee futures returns.

Quantifying consistency of Event Related Potentials across subjects based on single-trial topographic analysis

Introduction: Non-invasive brain imaging techniques often contrast experimental conditions across a cohort of participants, obfuscating distinctions in individual performance and brain mechanisms that are better characterised by the inter-trial variability. To overcome such limitations, we developed topographic analysis methods for single-trial EEG data [1]. So far this was typically based on time-frequency analysis of single-electrode data or single independent components. The method's efficacy is demonstrated for event-related responses to environmental sounds, hitherto studied at an average event-related potential (ERP) level. Methods: Nine healthy subjects participated to the experiment. Auditory meaningful sounds of common objects were used for a target detection task [2]. On each block, subjects were asked to discriminate target sounds, which were living or man-made auditory objects. Continuous 64-channel EEG was acquired during the task. Two datasets were considered for each subject including single-trial of the two conditions, living and man-made. The analysis comprised two steps. In the first part, a mixture of Gaussians analysis [3] provided representative topographies for each subject. In the second step, conditional probabilities for each Gaussian provided statistical inference on the structure of these topographies across trials, time, and experimental conditions. Similar analysis was conducted at group-level. Results: Results show that the occurrence of each map is structured in time and consistent across trials both at the single-subject and at group level. Conducting separate analyses of ERPs at single-subject and group levels, we could quantify the consistency of identified topographies and their time course of activation within and across participants as well as experimental conditions. A general agreement was found with previous analysis at average ERP level. Conclusions: This novel approach to single-trial analysis promises to have impact on several domains. In clinical research, it gives the possibility to statistically evaluate single-subject data, an essential tool for analysing patients with specific deficits and impairments and their deviation from normative standards. In cognitive neuroscience, it provides a novel tool for understanding behaviour and brain activity interdependencies at both single-subject and at group levels. In basic neurophysiology, it provides a new representation of ERPs and promises to cast light on the mechanisms of its generation and inter-individual variability.

An Introduction to parametric and non-parametric models for bivariate positive insurance claim severity distributions

We present a real data set of claims amounts where costs related to damage are recorded separately from those related to medical expenses. Only claims with positive costs are considered here. Two approaches to density estimation are presented: a classical parametric and a semi-parametric method, based on transformation kernel density estimation. We explore the data set with standard univariate methods. We also propose ways to select the bandwidth and transformation parameters in the univariate case based on Bayesian methods. We indicate how to compare the results of alternative methods both looking at the shape of the overall density domain and exploring the density estimates in the right tail.

Application of discriminant function analysis and change-point problem in dating volcanic ashes

The application of Discriminant function analysis (DFA) is not a new idea in the studyof tephrochrology. In this paper, DFA is applied to compositional datasets of twodifferent types of tephras from Mountain Ruapehu in New Zealand and MountainRainier in USA. The canonical variables from the analysis are further investigated witha statistical methodology of change-point problems in order to gain a betterunderstanding of the change in compositional pattern over time. Finally, a special caseof segmented regression has been proposed to model both the time of change and thechange in pattern. This model can be used to estimate the age for the unknown tephrasusing Bayesian statistical calibration

Probabilistic models to assist maintenance of multiple instruments

The paper discusses maintenance challenges of organisations with a huge number of devices and proposes the use of probabilistic models to assist monitoring and maintenance planning. The proposal assumes connectivity of instruments to report relevant features for monitoring. Also, the existence of enough historical registers with diagnosed breakdowns is required to make probabilistic models reliable and useful for predictive maintenance strategies based on them. Regular Markov models based on estimated failure and repair rates are proposed to calculate the availability of the instruments and Dynamic Bayesian Networks are proposed to model cause-effect relationships to trigger predictive maintenance services based on the influence between observed features and previously documented diagnostics

A Novel Breast Tissue Density Classification Methodology

It has been shown that the accuracy of mammographic abnormality detection methods is strongly dependent on the breast tissue characteristics, where a dense breast drastically reduces detection sensitivity. In addition, breast tissue density is widely accepted to be an important risk indicator for the development of breast cancer. Here, we describe the development of an automatic breast tissue classification methodology, which can be summarized in a number of distinct steps: 1) the segmentation of the breast area into fatty versus dense mammographic tissue; 2) the extraction of morphological and texture features from the segmented breast areas; and 3) the use of a Bayesian combination of a number of classifiers. The evaluation, based on a large number of cases from two different mammographic data sets, shows a strong correlation ( and 0.67 for the two data sets) between automatic and expert-based Breast Imaging Reporting and Data System mammographic density assessment

Proposta d'un model predictiu del risc d'ingrés hospitalari amb l'ús d'algoritmes bayesians

Model predictiu basat en xarxes bayesianes que permet identificar els pacients amb major risc d'ingrés a un hospital segons una sèrie d'atributs de dades demogràfiques i clíniques.

From Empirical Bayes to Full Bayes: Methods for Analysing Traffice Safety Data, October 25, 2004

Traffic safety engineers are among the early adopters of Bayesian statistical tools for analyzing crash data. As in many other areas of application, empirical Bayes methods were their first choice, perhaps because they represent an intuitively appealing, yet relatively easy to implement alternative to purely classical approaches. With the enormous progress in numerical methods made in recent years and with the availability of free, easy to use software that permits implementing a fully Bayesian approach, however, there is now ample justification to progress towards fully Bayesian analyses of crash data. The fully Bayesian approach, in particular as implemented via multi-level hierarchical models, has many advantages over the empirical Bayes approach. In a full Bayesian analysis, prior information and all available data are seamlessly integrated into posterior distributions on which practitioners can base their inferences. All uncertainties are thus accounted for in the analyses and there is no need to pre-process data to obtain Safety Performance Functions and other such prior estimates of the effect of covariates on the outcome of interest. In this slight, fully Bayesian methods may well be less costly to implement and may result in safety estimates with more realistic standard errors. In this manuscript, we present the full Bayesian approach to analyzing traffic safety data and focus on highlighting the differences between the empirical Bayes and the full Bayes approaches. We use an illustrative example to discuss a step-by-step Bayesian analysis of the data and to show some of the types of inferences that are possible within the full Bayesian framework.

From Empirical Bayes to Full Bayes: Methods for Analyzubg Traffic Safey Data, October 25, 2004

Traffic safety engineers are among the early adopters of Bayesian statistical tools for analyzing crash data. As in many other areas of application, empirical Bayes methods were their first choice, perhaps because they represent an intuitively appealing, yet relatively easy to implement alternative to purely classical approaches. With the enormous progress in numerical methods made in recent years and with the availability of free, easy to use software that permits implementing a fully Bayesian approach, however, there is now ample justification to progress towards fully Bayesian analyses of crash data. The fully Bayesian approach, in particular as implemented via multi-level hierarchical models, has many advantages over the empirical Bayes approach. In a full Bayesian analysis, prior information and all available data are seamlessly integrated into posterior distributions on which practitioners can base their inferences. All uncertainties are thus accounted for in the analyses and there is no need to pre-process data to obtain Safety Performance Functions and other such prior estimates of the effect of covariates on the outcome of interest. In this light, fully Bayesian methods may well be less costly to implement and may result in safety estimates with more realistic standard errors. In this manuscript, we present the full Bayesian approach to analyzing traffic safety data and focus on highlighting the differences between the empirical Bayes and the full Bayes approaches. We use an illustrative example to discuss a step-by-step Bayesian analysis of the data and to show some of the types of inferences that are possible within the full Bayesian framework.

Bayes linear spaces

Linear spaces consisting of σ-finite probability measures and infinite measures (improper priors and likelihood functions) are defined. The commutative group operation, called perturbation, is the updating given by Bayes theorem; the inverse operation is the Radon-Nikodym derivative. Bayes spaces of measures are sets of classes of proportional measures. In this framework, basic notions of mathematical statistics get a simple algebraic interpretation. For example, exponential families appear as affine subspaces with their sufficient statistics as a basis. Bayesian statistics, in particular some well-known properties of conjugated priors and likelihood functions, are revisited and slightly extended

Aineominaisuuksien mallinnustarkkuuden vaikutukset lämmönsiirtimien lämpöteknisessä simuloinnissa

Työssä on käsitelty fluidien aineominaisuuksien vaikutuksia paperikoneiden kuivatusosissa käytettävien lämmönsiirtimien lämpöteknisessä simuloinnissa. Pääkohteena selvitettiin kostean ilman ja veden fysikaalisien aineominaisuuksien mallinnustarkkuuden vaikutuksia lämpövirtaan lauhduttamattomissa ja lauhduttavissa tapauksissa. Asiaa tutkittiin tekemällä herkkyysanalyysi työssä kehitetyille termodynaamisille malleille. Perinteisen herkkyysanalyysin lisäksi herkkyyksiä tutkittiin myös Bayesiläisellä tilastoanalyysillä. Työssä käsiteltiin myös aineominaisuuksien käyttäytymistä ja mallintamista lämmönsiirtimissä. Kirjallisuudesta etsittiin aineominaisuusmallit, joilla kostean ilman ja veden fysikaalisia aineominaisuuksia voidaan kuvata riittävän tarkasti. Työssä havaittiin, että yksittäisistä aineominaisuuksista selkeästi suurimmat vaikutukset on ominaisentalpioiden mallinnuksen epätarkkuuksilla. Myös kaikkien aineominaisuuksien epätarkkuuksilla havaittiin olevan huomattavan suuret yhteisvaikutukset lämpövirran laskentatarkkuuteen. Viiden prosentin epätarkkuus kaikkien aineominaisuuksien mallinnuksessa johtaa 3 - 7 %:n epätarkkuuteen lämpövirran laskennassa. Näin ollen kaikkien aineominaisuuksien mallintamiseen tulee kiinnittää huomiota.

Impact of a cis-associated gene expression SNP on chromosome 20q11.22 on bipolar disorder susceptibility, hippocampal structure and cognitive performance.

BackgroundBipolar disorder is a highly heritable polygenic disorder. Recent enrichment analyses suggest that there may be true risk variants for bipolar disorder in the expression quantitative trait loci (eQTL) in the brain.AimsWe sought to assess the impact of eQTL variants on bipolar disorder risk by combining data from both bipolar disorder genome-wide association studies (GWAS) and brain eQTL.MethodTo detect single nucleotide polymorphisms (SNPs) that influence expression levels of genes associated with bipolar disorder, we jointly analysed data from a bipolar disorder GWAS (7481 cases and 9250 controls) and a genome-wide brain (cortical) eQTL (193 healthy controls) using a Bayesian statistical method, with independent follow-up replications. The identified risk SNP was then further tested for association with hippocampal volume (n = 5775) and cognitive performance (n = 342) among healthy individuals.ResultsIntegrative analysis revealed a significant association between a brain eQTL rs6088662 on chromosome 20q11.22 and bipolar disorder (log Bayes factor = 5.48; bipolar disorder P = 5.85×10(-5)). Follow-up studies across multiple independent samples confirmed the association of the risk SNP (rs6088662) with gene expression and bipolar disorder susceptibility (P = 3.54×10(-8)). Further exploratory analysis revealed that rs6088662 is also associated with hippocampal volume and cognitive performance in healthy individuals.ConclusionsOur findings suggest that 20q11.22 is likely a risk region for bipolar disorder; they also highlight the informative value of integrating functional annotation of genetic variants for gene expression in advancing our understanding of the biological basis underlying complex disorders, such as bipolar disorder.

Social differences in avoidable mortality between small areas of 15 European cities: an ecological study

Health and inequalities in health among inhabitants of European cities are of major importance for European public health and there is great interest in how different health care systems in Europe perform in the reduction of health inequalities. However, evidence on the spatial distribution of cause-specific mortality across neighbourhoods of European cities is scarce. This study presents maps of avoidable mortality in European cities and analyses differences in avoidable mortality between neighbourhoods with different levels of deprivation. Methods: We determined the level of mortality from 14 avoidable causes of death for each neighbourhood of 15 large cities in different European regions. To address the problems associated with Standardised Mortality Ratios for small areas we smooth them using the Bayesian model proposed by Besag, York and Mollié. Ecological regression analysis was used to assess the association between social deprivation and mortality. Results: Mortality from avoidable causes of death is higher in deprived neighbourhoods and mortality rate ratios between areas with different levels of deprivation differ between gender and cities. In most cases rate ratios are lower among women. While Eastern and Southern European cities show higher levels of avoidable mortality, the association of mortality with social deprivation tends to be higher in Northern and lower in Southern Europe. Conclusions: There are marked differences in the level of avoidable mortality between neighbourhoods of European cities and the level of avoidable mortality is associated with social deprivation. There is no systematic difference in the magnitude of this association between European cities or regions. Spatial patterns of avoidable mortality across small city areas can point to possible local problems and specific strategies to reduce health inequality which is important for the development of urban areas and the well-being of their inhabitants

Panel Data Stochastic Convergence Analysis of the Mexican Regions

The stochastic convergence amongst Mexican Federal entities is analyzed in panel data framework. The joint consideration of cross-section dependence and multiple structural breaks is required to ensure that the statistical inference is based on statistics with good statistical properties. Once these features are accounted for, evidence in favour of stochastic convergence is found. Since stochastic convergence is a necessary, yet insufficient condition for convergence as predicted by economic growth models, the paper also investigates whether-convergence process has taken place. We found that the Mexican states have followed either heterogeneous convergence patterns or divergence process throughout the analyzed period.