Modelling dependence in a ratemaking procedure with multivariate Poisson regression models

When actuaries face with the problem of pricing an insurance contract that contains different types of coverage, such as a motor insurance or homeowner's insurance policy, they usually assume that types of claim are independent. However, this assumption may not be realistic: several studies have shown that there is a positive correlation between types of claim. Here we introduce different regression models in order to relax the independence assumption, including zero-inflated models to account for excess of zeros and overdispersion. These models have been largely ignored to multivariate Poisson date, mainly because of their computational di±culties. Bayesian inference based on MCMC helps to solve this problem (and also lets us derive, for several quantities of interest, posterior summaries to account for uncertainty). Finally, these models are applied to an automobile insurance claims database with three different types of claims. We analyse the consequences for pure and loaded premiums when the independence assumption is relaxed by using different multivariate Poisson regression models and their zero-inflated versions.

An R Library for Compositional Data Analysis in Archaeometry

Compositional data naturally arises from the scientific analysis of the chemicalcomposition of archaeological material such as ceramic and glass artefacts. Data of thistype can be explored using a variety of techniques, from standard multivariate methodssuch as principal components analysis and cluster analysis, to methods based upon theuse of log-ratios. The general aim is to identify groups of chemically similar artefactsthat could potentially be used to answer questions of provenance.This paper will demonstrate work in progress on the development of a documentedlibrary of methods, implemented using the statistical package R, for the analysis ofcompositional data. R is an open source package that makes available very powerfulstatistical facilities at no cost. We aim to show how, with the aid of statistical softwaresuch as R, traditional exploratory multivariate analysis can easily be used alongside, orin combination with, specialist techniques of compositional data analysis.The library has been developed from a core of basic R functionality, together withpurpose-written routines arising from our own research (for example that reported atCoDaWork'03). In addition, we have included other appropriate publicly availabletechniques and libraries that have been implemented in R by other authors. Availablefunctions range from standard multivariate techniques through to various approaches tolog-ratio analysis and zero replacement. We also discuss and demonstrate a smallselection of relatively new techniques that have hitherto been little-used inarchaeometric applications involving compositional data. The application of the libraryto the analysis of data arising in archaeometry will be demonstrated; results fromdifferent analyses will be compared; and the utility of the various methods discussed

Diagnosing Patients Combining Principal Components Analysis and Case Based Reasoning

This paper addresses the application of a PCA analysis on categorical data prior to diagnose a patients data set using a Case-Based Reasoning (CBR) system. The particularity is that the standard PCA techniques are designed to deal with numerical attributes, but our medical data set contains many categorical data and alternative methods as RS-PCA are required. Thus, we propose to hybridize RS-PCA (Regular Simplex PCA) and a simple CBR. Results show how the hybrid system produces similar results when diagnosing a medical data set, that the ones obtained when using the original attributes. These results are quite promising since they allow to diagnose with less computation effort and memory storage

Elemental analysis of New Zealand Nephrite Jade (pounamu) for forensic purposes

Pounamu (NZ jade), or nephrite, is a protected mineral in its natural form following thetransfer of ownership back to Ngai Tahu under the Ngai Tahu (Pounamu Vesting) Act 1997.Any theft of nephrite is prosecutable under the Crimes Act 1961. Scientific evidence isessential in cases where origin is disputed. A robust method for discrimination of thismaterial through the use of elemental analysis and compositional data analysis is required.Initial studies have characterised the variability within a given nephrite source. This hasincluded investigation of both in situ outcrops and alluvial material. Methods for thediscrimination of two geographically close nephrite sources are being developed.Key Words: forensic, jade, nephrite, laser ablation, inductively coupled plasma massspectrometry, multivariate analysis, elemental analysis, compositional data analysis

Some last thoughts on compositional data analysis

One of the disadvantages of old age is that there is more past than future: this,however, may be turned into an advantage if the wealth of experience and, hopefully,wisdom gained in the past can be reflected upon and throw some light on possiblefuture trends. To an extent, then, this talk is necessarily personal, certainly nostalgic,but also self critical and inquisitive about our understanding of the discipline ofstatistics. A number of almost philosophical themes will run through the talk: searchfor appropriate modelling in relation to the real problem envisaged, emphasis onsensible balances between simplicity and complexity, the relative roles of theory andpractice, the nature of communication of inferential ideas to the statistical layman, theinter-related roles of teaching, consultation and research. A list of keywords might be:identification of sample space and its mathematical structure, choices betweentransform and stay, the role of parametric modelling, the role of a sample spacemetric, the underused hypothesis lattice, the nature of compositional change,particularly in relation to the modelling of processes. While the main theme will berelevance to compositional data analysis we shall point to substantial implications forgeneral multivariate analysis arising from experience of the development ofcompositional data analysis…

Diagnosing Patients Combining Principal Components Analysis and Case Based Reasoning

This paper addresses the application of a PCA analysis on categorical data prior to diagnose a patients data set using a Case-Based Reasoning (CBR) system. The particularity is that the standard PCA techniques are designed to deal with numerical attributes, but our medical data set contains many categorical data and alternative methods as RS-PCA are required. Thus, we propose to hybridize RS-PCA (Regular Simplex PCA) and a simple CBR. Results show how the hybrid system produces similar results when diagnosing a medical data set, that the ones obtained when using the original attributes. These results are quite promising since they allow to diagnose with less computation effort and memory storage

Analysis of sensitivity with respect to a compositional parameter : A comment on 'Local sensitivity analysis for compositional data with application to soil texture in hydrologic modelling' by L. Loosvelt and co-authors

A comment about the article “Local sensitivity analysis for compositional data with application to soil texture in hydrologic modelling” writen by L. Loosvelt and co-authors. The present comment is centered in three specific points. The first one is related to the fact that the authors avoid the use of ilr-coordinates. The second one refers to some generalization of sensitivity analysis when input parameters are compositional. The third tries to show that the role of the Dirichlet distribution in the sensitivity analysis is irrelevant

Subtypes of adolescents with substance use disorders and psychiatric comorbidity using cluster and discriminant analysis of MMPI-A profiles

The main aim of this study was to replicate and extend previous results on subtypes of adolescents with substance use disorders (SUD), according to their Minnesota Multiphasic Personality Inventory for adolescents (MMPI-A) profiles. Sixty patients with SUD and psychiatric comorbidity (41.7% male, mean age = 15.9 years old) completed the MMPI-A, the Teen Addiction Severity Index (T-ASI), the Child Behaviour Checklist (CBCL), and were interviewed in order to determine DSMIV diagnoses and level of substance use. Mean MMPI-A personality profile showed moderate peaks in Psychopathic Deviate, Depression and Hysteria scales. Hierarchical cluster analysis revealed four profiles (acting-out, 35% of the sample; disorganized-conflictive, 15%; normative-impulsive, 15%; and deceptive-concealed, 35%). External correlates were found between cluster 1, CBCL externalizing symptoms at a clinical level and conduct disorders, and between cluster 2 and mixed CBCL internalized/externalized symptoms at a clinical level. Discriminant analysis showed that Depression, Psychopathic Deviate and Psychasthenia MMPI-A scales correctly classified 90% of the patients into the clusters obtained.

Insomnia in hospitalised patients with advanced disease: prevalence and related factors

Objectiu: Avaluar prevalença i factors associats a l’insomni en pacients ingressats a una unitat de cures pal•liatives. Mètode: Entrevista estructurada a pacients consecutivament ingressats avaluant l’insomni a través de l’escala Sleep Disturbance Scale, i els factors físics, psicològics i ambientals potencialment associats. Resultats principals: El 47% presentà insomni moderat a sever. Els factors potencialment associats més prevalents foren dolor, distrés psicològic, rumiacions nocturnes i factors ambientals. En l’anàlisi multivariada, rumiacions nocturnes i somnolència diürna (relació inversa) es relacionaren amb insomni moderat a sever. Conclusions: L’insomni és un símptoma prevalent relacionat amb rumiacions nocturnes i absència de somnolència diürna.

Valor pronóstico de la densidad microvascular y de la expresión del VEGF, EGFR y HIF-1 en pacientes con cáncer de cérvix localmente avanzado tratados con quimioradiación

Introducción: Colombia cuenta con poca información sobre el comportamiento del cáncer, no obstante, el carcinoma de cuello uterino representa la segunda causa de muerte por la enfermedad entre las mujeres de nuestro entorno. El patrón epidemiológico de la enfermedad es preocupante porque los estados localmente avanzados constituyen el estado más frecuente al momento del diagnóstico y la mortalidad siendo bastante alta a pesar de la presencia de un programa de cribado organizado. Objetivo: Describir el valor pronóstico de la densidad microvascular (DMV) y de la expresión proteica de varios genes relacionados con la supervivencia y proliferación del cáncer de cérvix localmente avanzado en un grupo de mujeres tratadas con quimioradiación y braquiterapia intracavitaria. Se estimaron la tasa de respuesta global (TRG), la supervivencia libre de progresión (SLP) y la supervivencia global (SG). Resultados: Se incluyeron 61 mujeres con una edad media de 52 ± 10 años; todas tenían diagnóstico de cáncer de cérvix localmente avanzado (IIA 2.3%/IIB 47.5%/IIIA 4.9%/IIIB 37.7%/IVA 3.3%/no definido 3.3%), con un volumen tumoral promedio de 6.4cm (DE ± 1.8cm) e infección por VPH en 46% de los casos; 58 sujetos (95%) tenían un patrón escamoso, dos fueron adenocarcinomas y &50% presentaba neoplasias moderada o pobremente diferenciadas. Todas fueron tratadas con quimioradiación (interrupción transitoria en teleterapia por toxicidad y otras causas en 19% y 21.4%, respectivamente/media de ciclos de platino concomitante 4.8 series ± 1.0) y braquiterapia (77% completaron el tratamiento intracavitario). La mediana para la SLP y global fue de 6.6 meses (r, 4.0-9.1) y 30 meses (r, 11-48), respectivamente. Ninguna de las variables tuvo un efecto positivo sobre la SLP, mientras el análisis multivariado demostró que los niveles de expresión del VEGF (P=0.026), EGFR (P=0.030), y el volumen tumoral menor de 6 cm (P=0.02) influyeron positivamente sobre éste desenlace. Conclusión: Existe una influencia positiva sobre el pronóstico, de la tipificación en el cáncer de cérvix localmente avanzado tratado con quimioradiación basada en platino.

Intra-Party Democracy and middle-level elites in Spain

The gradual implementation of new, more participatory and thus, more democratic mechanisms of intra-party decision-making has been pointed out by several party politics scholars. This phenomenon has been studied as the party elite’s reactions to a widespread trend in Western countries: the party membership decline. Spain is still a deviant case in both the party membership decline trend, and with regards to the introduction of more participatory and democratic decision-making mechanisms. However, the paper point out that support for intra-party democracy is quite widespread within Spanish party middle elites (party delegates). That is why the aim of this paper is to explain which factors are underpinning the supports for intra-party democracy amongst Spanish party delegates. After conducting a multivariate analysis, the results show that ideology, the involvement in intra-party experiences and the degree of pragmatism, amongst others, are factors strongly associated with the support for intraparty democracy in Spanish party middle elites.

Simultaneous estimation of indirect and interaction effects using structural equation models

Interaction effects are usually modeled by means of moderated regression analysis. Structural equation models with non-linear constraints make it possible to estimate interaction effects while correcting formeasurement error. From the various specifications, Jöreskog and Yang's(1996, 1998), likely the most parsimonious, has been chosen and further simplified. Up to now, only direct effects have been specified, thus wasting much of the capability of the structural equation approach. This paper presents and discusses an extension of Jöreskog and Yang's specification that can handle direct, indirect and interaction effects simultaneously. The model is illustrated by a study of the effects of an interactive style of use of budgets on both company innovation and performance

Compositional hypotheses of subcompositional stability and specific perturbation change and their testing

In standard multivariate statistical analysis common hypotheses of interest concern changes in mean vectors and subvectors. In compositional data analysis it is now well established that compositional change is most readily described in terms of the simplicial operation of perturbation and that subcompositions replace the marginal concept of subvectors. To motivate the statistical developments of this paper we present two challenging compositional problems from food production processes.Against this background the relevance of perturbations and subcompositions can beclearly seen. Moreover we can identify a number of hypotheses of interest involvingthe specification of particular perturbations or differences between perturbations and also hypotheses of subcompositional stability. We identify the two problems as being the counterpart of the analysis of paired comparison or split plot experiments and of separate sample comparative experiments in the jargon of standard multivariate analysis. We then develop appropriate estimation and testing procedures for a complete lattice of relevant compositional hypotheses

Improving Clustering Algorithms for Image Segmentation using Contour and Region Information

In image segmentation, clustering algorithms are very popular because they are intuitive and, some of them, easy to implement. For instance, the k-means is one of the most used in the literature, and many authors successfully compare their new proposal with the results achieved by the k-means. However, it is well known that clustering image segmentation has many problems. For instance, the number of regions of the image has to be known a priori, as well as different initial seed placement (initial clusters) could produce different segmentation results. Most of these algorithms could be slightly improved by considering the coordinates of the image as features in the clustering process (to take spatial region information into account). In this paper we propose a significant improvement of clustering algorithms for image segmentation. The method is qualitatively and quantitative evaluated over a set of synthetic and real images, and compared with classical clustering approaches. Results demonstrate the validity of this new approach

Unusual voltage sag event detection in power systems

Three multivariate statistical tools (principal component analysis, factor analysis, analysis discriminant) have been tested to characterize and model the sags registered in distribution substations. Those models use several features to represent the magnitude, duration and unbalanced grade of sags. They have been obtained from voltage and current waveforms. The techniques are tested and compared using 69 registers of sags. The advantages and drawbacks of each technique are listed