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.

A tale of two logits, compositional data analysis and zero observations

The application of compositional data analysis through log ratio trans-formations corresponds to a multinomial logit model for the shares themselves.This model is characterized by the property of Independence of Irrelevant Alter-natives (IIA). IIA states that the odds ratio in this case the ratio of shares is invariant to the addition or deletion of outcomes to the problem. It is exactlythis invariance of the ratio that underlies the commonly used zero replacementprocedure in compositional data analysis. In this paper we investigate using thenested logit model that does not embody IIA and an associated zero replacementprocedure and compare its performance with that of the more usual approach ofusing the multinomial logit model. Our comparisons exploit a data set that com-bines voting data by electoral division with corresponding census data for eachdivision for the 2001 Federal election in Australia

Multiuser detection in a dynamic environment — Part II: Joint user identification and parameter estimation

The problem of jointly estimating the number, the identities, and the data of active users in a time-varying multiuser environment was examined in a companion paper (IEEE Trans. Information Theory, vol. 53, no. 9, September 2007), at whose core was the use of the theory of finite random sets on countable spaces. Here we extend that theory to encompass the more general problem of estimating unknown continuous parameters of the active-user signals. This problem is solved here by applying the theory of random finite sets constructed on hybrid spaces. We doso deriving Bayesian recursions that describe the evolution withtime of a posteriori densities of the unknown parameters and data.Unlike in the above cited paper, wherein one could evaluate theexact multiuser set posterior density, here the continuous-parameter Bayesian recursions do not admit closed-form expressions. To circumvent this difficulty, we develop numerical approximationsfor the receivers that are based on Sequential Monte Carlo (SMC)methods (“particle filtering”). Simulation results, referring to acode-divisin multiple-access (CDMA) system, are presented toillustrate the theory.

Sexual foraging segregation in South American sea lions increases during the pre-breeding period in the La Plata River plume

Stable carbon and nitrogen isotopes in skin and bone of South American sea lions from Brazil and Uruguay were analysed to test the hypothesis that trophic overlap between the sexes is lower during the pre-breeding season than throughout the rest of the year. The isotopic values of skin and bone were used to infer the trophic relationships between the sexes during the pre-breeding period and year round, respectively. Prey species were also analysed to establish a baseline necessary for interpreting the stable isotope ratios of skin and bone. Standard ellipse areas, estimated using Bayesian inference in the SIBER routine of the SIAR package in R, suggested that males and females used a wide diversity of foraging strategies throughout the year and that no differences existed between the sexes. However, the diversity of foraging strategies was largely reduced during the pre-breeding period, with all the individuals of each sex adopting similar strategies, but with the two sexes differing considerably in stable isotope values and the ellipse areas of males and females not overlapping at all. Nevertheless, the results revealed a general increase in the consumption of pelagic prey by both sexes during the pre-breeding period. The progressive crowding of individuals in the areas surrounding the breeding rookeries during the pre-breeding period could lead to an increase in the local population density, which could explain the above reported changes.

Productive efficiency and regulatory reform: The case of vehicle inspection services

Measuring productive efficiency provides information on the likely effects of regulatory reform. We present a Data Envelopment Analysis (DEA) of a sample of 38 vehicle inspection units under a concession regime, between the years 2000 and 2004. The differences in efficiency scores show the potential technical efficiency benefit of introducing some form of incentive regulation or of progressing towards liberalization. We also compute scale efficiency scores, showing that only units in territories with very low population density operate at a sub-optimal scale. Among those that operate at an optimal scale, there are significant differences in size; the largest ones operate in territories with the highest population density. This suggests that the introduction of new units in the most densely populated territories (a likely effect of some form of liberalization) would not be detrimental in terms of scale efficiency. We also find that inspection units belonging to a large, diversified firm show higher technical efficiency, reflecting economies of scale or scope at the firm level. Finally, we show that between 2002 and 2004, a period of high regulatory uncertainty in the sample’s region, technical change was almost zero. Regulatory reform should take due account of scale and diversification effects, while at the same time avoiding regulatory uncertainty.

Risc i efectes a la salut de la desinfecció de l'aigua: una aproximació a la relació entre els subproductes de la cloració de l'aigua i el càncer colorectal

Aquest projecte es centra en l'estimació del risc ambiental que suposen els subproductes de la desinfecció de l'aigua (amb els trihalometans com a marcadors) per a la incidència de càncer colorectal. Això s'estudia mesurant les exposicions i usos de l'aigua d'un grup de casos i controls a diversos llocs de l'Estat Espanyol, comparant els nivells de trihalometans d'aquestes ciutats, estimant-ne l'absorció i, per últim, calculant-ne el risc mitjançant la odds ratio (probabilitat de patir la malaltia dividit per la probabilitat de no desenvolupar-la). Per assolir-ho és vital l'acurada estimació de l'exposició de la població a aquests productes, ja que, tot i que l'aigua de xarxa és una exposició ubiqua i, per tant, ambiental, hi ha persones més exposades que d'altres. Comparant els nivells de trihalometans arreu d'Espanya, la qualitat de l'aigua de consum varia molt als diferents llocs d'estudi, essent Barcelona la de pitjor qualitat ambiental i Guipúscoa la millor. La primera estimació al risc (l'estudi oficial, dut a terme pel Centre de Recerca en Epidemiologia Ambiental, no està conclòs) dóna positiu en la majoria d'exposicions, amb l'excepció de l'assistència a piscines durant l'hivern.

Prevalencia de enfermedad arterial periférica en pacientes hipertensos de la población de Paterna mayores de 15 años mediante determinación del índice tobillo-brazo.

Objectius: Conèixer la prevalença de malaltia arterial perifèrica en malalts hipertensos. Material y Métodes. Estudi descriptiu transversal en una mostra de 300 persones hipertenses majors de 15 anys obtingudes al atzar, ateses al centre de Salut de Paterna. Resultats. La prevalença fou d´un 16,7% determinada per un IMB patològic (&0.90). En l´anàlisi multifactorial sols va resultar significativa la variable sexe odds ratio (OR = 2,43) Conclusions. Es necesari prendre conciència de la importància de la MAP per la seua forta capacitat predictora de mal pronòstic a curt-mitjà termini. El IMB es una eina útil, senzilla i barata.

Efectividad de la Vacuna Neumocócica Polisacarida Polivalente en la prevención de enfermedad invasiva por neumococo en personas mayores de 60 años.

INTRODUCCIÓ La vacuna pneumocòccica polisacàrida polivalent (VPP-23) es recomana en ancians i persones d'alt risc. No obstant això, la seva efectivitat en la prevenció d'infeccions per pneumococ és controvertida. Aquest estudi avalua l'efectivitat de la vacunació en la prevenció de malaltia pneumocòccica invasiva (MPI) en persones majors de 60 anys. METODOLOGIA Estudi poblacional de casos i controls en el qual es van incloure 88 pacients majors de 60 anys amb MPI (Pneumònia bacteriana, meningitis o sepsis) confirmada per laboratori, i 176 controls que van ser assignats segons Centre d'Atenció Primària, edat, sexe i nivell de risc. La regressió logística condicional va ser usada per a estimar la odds ràtio (OR) segons cada condició clínica. L'efectivitat de la vacuna va ser estimada segons (1-OR) x100. RESULTATS L'efectivitat de la vacunació pneumocòccica va ser significativament mes baixa en casos que en controls (38.6% vs 59.1%; p=0.002). L'efectivitat ajustada va anar de 72% (OR: 0.28; 95% CI: 0.15-0.54) per a MPI i 77% (OR: 0.23; 95% CI: 0.08-0.60) per als serotips vacunals inclosos en la VPP-23. La vacunació va ser efectiva per a l'MPI tant en el grup d'edat de 60-79 anys (OR 0.32; 95% CI: 0.14-0.74) com en el grup de 80 anys o mes (OR: 0.29; 95% CI: 0.09-0.91). L'efectivitat vacunal va ser estadísticament significativa en persones d'alt risc inmunocompetents (OR: 0.29; 95% CI: 0.11-0.79) així com en persones inmunocompromeses (OR: 0.12; 95% CI: 0.03- 0.53). CONCLUSIÓ Aquestes troballes confirmen l'efectivitat de la vacuna pneumocòccia polisacàrida polivalent (VPP-23) en l'MPI, i el benefici en la prevenció d'infecció invasiva en persones de risc i en ancians.