Estimation of parametric and nonparametric models for univariate claim severity distributions - an approach using R

This paper presents an analysis of motor vehicle insurance claims relating to vehicle damage and to associated medical expenses. We use univariate severity distributions estimated with parametric and non-parametric methods. The methods are implemented using the statistical package R. Parametric analysis is limited to estimation of normal and lognormal distributions for each of the two claim types. The nonparametric analysis presented involves kernel density estimation. We illustrate the benefits of applying transformations to data prior to employing kernel based methods. We use a log-transformation and an optimal transformation amongst a class of transformations that produces symmetry in the data. The central aim of this paper is to provide educators with material that can be used in the classroom to teach statistical estimation methods, goodness of fit analysis and importantly statistical computing in the context of insurance and risk management. To this end, we have included in the Appendix of this paper all the R code that has been used in the analysis so that readers, both students and educators, can fully explore the techniques described

Policy evaluation and uncertainty about the effects of oil prices on economic activity

This paper addresses the issue of policy evaluation in a context in which policymakers are uncertain about the effects of oil prices on economic performance. I consider models of the economy inspired by Solow (1980), Blanchard and Gali (2007), Kim and Loungani (1992) and Hamilton (1983, 2005), which incorporate different assumptions on the channels through which oil prices have an impact on economic activity. I first study the characteristics of the model space and I analyze the likelihood of the different specifications. I show that the existence of plausible alternative representations of the economy forces the policymaker to face the problem of model uncertainty. Then, I use the Bayesian approach proposed by Brock, Durlauf and West (2003, 2007) and the minimax approach developed by Hansen and Sargent (2008) to integrate this form of uncertainty into policy evaluation. I find that, in the environment under analysis, the standard Taylor rule is outperformed under a number of criteria by alternative simple rules in which policymakers introduce persistence in the policy instrument and respond to changes in the real price of oil.

What do we gain from simplicity versus complexity in species distribution models?

Species distribution models (SDMs) are widely used to explain and predict species ranges and environmental niches. They are most commonly constructed by inferring species' occurrence-environment relationships using statistical and machine-learning methods. The variety of methods that can be used to construct SDMs (e.g. generalized linear/additive models, tree-based models, maximum entropy, etc.), and the variety of ways that such models can be implemented, permits substantial flexibility in SDM complexity. Building models with an appropriate amount of complexity for the study objectives is critical for robust inference. We characterize complexity as the shape of the inferred occurrence-environment relationships and the number of parameters used to describe them, and search for insights into whether additional complexity is informative or superfluous. By building 'under fit' models, having insufficient flexibility to describe observed occurrence-environment relationships, we risk misunderstanding the factors shaping species distributions. By building 'over fit' models, with excessive flexibility, we risk inadvertently ascribing pattern to noise or building opaque models. However, model selection can be challenging, especially when comparing models constructed under different modeling approaches. Here we argue for a more pragmatic approach: researchers should constrain the complexity of their models based on study objective, attributes of the data, and an understanding of how these interact with the underlying biological processes. We discuss guidelines for balancing under fitting with over fitting and consequently how complexity affects decisions made during model building. Although some generalities are possible, our discussion reflects differences in opinions that favor simpler versus more complex models. We conclude that combining insights from both simple and complex SDM building approaches best advances our knowledge of current and future species ranges.

Participation of interleukin-5, interleukin-8 and leukotriene B4 in eosinophil accumulation in two different experimental models

There are several experimental models describing in vivo eosinophil (EO) migration, including ip injection of a large volume of saline (SAL) or Sephadex beads (SEP). The aim of this study was to investigate the mechanisms involved in the EO migration in these two models. Two consecutive injections of SAL given 48 hr apart, induced a selective recruitment of EO into peritoneal cavity of rats, which peaked 48 hr after the last injection. SEP, when injected ip, promoted EO accumulation in rats. The phenomenom was dose-related and peaked 48 hr after SEP injection. To investigate the mediators involved in this process we showed that BW A4C, MK 886 and dexamethasone (DXA) inhibited the EO migration induced by SAL and SEP. To investigate the source of the EO chemotactic factor we showed that mast cells, macrophages (MO), but not lymphocytes, incubated in vitro in presence of SAL released a factor which induced EO migration. With SEP, only mast cells release a factor that induced EO migration, which was inhibited by BW A4C, MK 886 and DXA. Furthermore, the chemotactic activity of SAL-stimulated mast cells was inhibited by antisera against IL-5 and IL-8 (interleukin). SAL-stimulated MO were only inhibited by anti-IL-8 antibodies as well SEP-stimulated mast cells. These results suggest that the EO migration induced by SAL may be dependent on resident mast cells and MO and mediated by LTB4, IL-5 and IL-8. SEP-induced EO migration was dependent on mast cells and may be mediated by LTB4 and IL-8. Furthermore, IL-5 and IL-8 induced EO migration, which was also dependent on resident cells and mediated by LTB4 . In conclusion, EO migration induced by SAL is dependent on mast cells and MO, whereas that induced by SEP is dependent on mast cells alone. Stimulated mast cells release LTB4, IL-5 and IL-8 while MO release LTB4 and IL-8. The IL-5 and IL-8 release by the SAL or SEP-stimulated resident cells may act in an autocrine fashion, thus potentiating LTB4 release.

Role of eosinophilic airway inflammation in models of asthma

Eosinophils play a central role in the establishment and outcome of bronchial inflammation in asthma. Animal models of allergy are useful to answer questions related to mechanisms of allergic inflammation. We have used models of sensitized and boosted guinea pigs to investigate the nature of bronchial inflammation in allergic conditions. These animals develop marked bronchial infiltration composed mainly of CD4+ T-lymphocytes and eosinophils. Further provocation with antigen leads to degranulation of eosinophils and ulceration of the bronchial mucosa. Eosinophils are the first cells to increase in numbers in the mucosa after antigen challenge and depend on the expression of alpha 4 integrin to adhere to the vascular endothelium and transmigrate to the mucosa. Blockage of alpha4 integrin expression with specific antibody prevents not only the transmigration of eosinophils but also the development of bronchial hyperresponsiveness (BHR) to agonists in sensitized and challenged animals, clearly suggesting a role for this cell type in this altered functional state. Moreover, introduction of antibody against Major Basic Protein into the airways also prevents the development of BHR in similar model. BHR can also be suppressed by the use of FK506, an immunosuppressor that reduces in almost 100% the infiltration of eosinophils into the bronchi of allergic animals. These data support the concept that eosinophil is the most important pro-inflammatory factor in bronchial inflammation associated with allergy.

Indirect likelihood inference

Given a sample from a fully specified parametric model, let Zn be a given finite-dimensional statistic - for example, an initial estimator or a set of sample moments. We propose to (re-)estimate the parameters of the model by maximizing the likelihood of Zn. We call this the maximum indirect likelihood (MIL) estimator. We also propose a computationally tractable Bayesian version of the estimator which we refer to as a Bayesian Indirect Likelihood (BIL) estimator. In most cases, the density of the statistic will be of unknown form, and we develop simulated versions of the MIL and BIL estimators. We show that the indirect likelihood estimators are consistent and asymptotically normally distributed, with the same asymptotic variance as that of the corresponding efficient two-step GMM estimator based on the same statistic. However, our likelihood-based estimators, by taking into account the full finite-sample distribution of the statistic, are higher order efficient relative to GMM-type estimators. Furthermore, in many cases they enjoy a bias reduction property similar to that of the indirect inference estimator. Monte Carlo results for a number of applications including dynamic and nonlinear panel data models, a structural auction model and two DSGE models show that the proposed estimators indeed have attractive finite sample properties.

Industrial Location and Space: New Insights

This paper tries to resolve some of the main shortcomings in the empirical literature of location decisions for new plants, i.e. spatial effects and overdispersion. Spatial effects are omnipresent, being a source of overdispersion in the data as well as a factor shaping the functional relationship between the variables that explain a firm’s location decisions. Using Count Data models, empirical researchers have dealt with overdispersion and excess zeros by developments of the Poisson regression model. This study aims to take this a step further, by adopting Bayesian methods and models in order to tackle the excess of zeros, spatial and non-spatial overdispersion and spatial dependence simultaneously. Data for Catalonia is used and location determinants are analysed to that end. The results show that spatial effects are determinant. Additionally, overdispersion is descomposed into an unstructured iid effect and a spatially structured effect. Keywords: Bayesian Analysis, Spatial Models, Firm Location. JEL Classification: C11, C21, R30.

Mixture of bivariate Poisson regression models with an application to insurance

In a recent paper Bermúdez [2009] used bivariate Poisson regression models for ratemaking in car insurance, and included zero-inflated models to account for the excess of zeros and the overdispersion in the data set. In the present paper, we revisit this model in order to consider alternatives. We propose a 2-finite mixture of bivariate Poisson regression models to demonstrate that the overdispersion in the data requires more structure if it is to be taken into account, and that a simple zero-inflated bivariate Poisson model does not suffice. At the same time, we show that a finite mixture of bivariate Poisson regression models embraces zero-inflated bivariate Poisson regression models as a special case. Additionally, we describe a model in which the mixing proportions are dependent on covariates when modelling the way in which each individual belongs to a separate cluster. Finally, an EM algorithm is provided in order to ensure the models’ ease-of-fit. These models are applied to the same automobile insurance claims data set as used in Bermúdez [2009] and it is shown that the modelling of the data set can be improved considerably.

Climate-based empirical models show biased predictions of butterfly communities along environmental gradients

A better understanding of the factors that mould ecological community structure is required to accurately predict community composition and to anticipate threats to ecosystems due to global changes. We tested how well stacked climate-based species distribution models (S-SDMs) could predict butterfly communities in a mountain region. It has been suggested that climate is the main force driving butterfly distribution and community structure in mountain environments, and that, as a consequence, climate-based S-SDMs should yield unbiased predictions. In contrast to this expectation, at lower altitudes, climate-based S-SDMs overpredicted butterfly species richness at sites with low plant species richness and underpredicted species richness at sites with high plant species richness. According to two indices of composition accuracy, the Sorensen index and a matching coefficient considering both absences and presences, S-SDMs were more accurate in plant-rich grasslands. Butterflies display strong and often specialised trophic interactions with plants. At lower altitudes, where land use is more intense, considering climate alone without accounting for land use influences on grassland plant richness leads to erroneous predictions of butterfly presences and absences. In contrast, at higher altitudes, where climate is the main force filtering communities, there were fewer differences between observed and predicted butterfly richness. At high altitudes, even if stochastic processes decrease the accuracy of predictions of presence, climate-based S-SDMs are able to better filter out butterfly species that are unable to cope with severe climatic conditions, providing more accurate predictions of absences. Our results suggest that predictions should account for plants in disturbed habitats at lower altitudes but that stochastic processes and heterogeneity at high altitudes may limit prediction success of climate-based S-SDMs.

Stochastic time-concentration activity models for cytotoxicity in 3D brain cell cultures.

BACKGROUND: In vitro aggregating brain cell cultures containing all types of brain cells have been shown to be useful for neurotoxicological investigations. The cultures are used for the detection of nervous system-specific effects of compounds by measuring multiple endpoints, including changes in enzyme activities. Concentration-dependent neurotoxicity is determined at several time points. METHODS: A Markov model was set up to describe the dynamics of brain cell populations exposed to potentially neurotoxic compounds. Brain cells were assumed to be either in a healthy or stressed state, with only stressed cells being susceptible to cell death. Cells may have switched between these states or died with concentration-dependent transition rates. Since cell numbers were not directly measurable, intracellular lactate dehydrogenase (LDH) activity was used as a surrogate. Assuming that changes in cell numbers are proportional to changes in intracellular LDH activity, stochastic enzyme activity models were derived. Maximum likelihood and least squares regression techniques were applied for estimation of the transition rates. Likelihood ratio tests were performed to test hypotheses about the transition rates. Simulation studies were used to investigate the performance of the transition rate estimators and to analyze the error rates of the likelihood ratio tests. The stochastic time-concentration activity model was applied to intracellular LDH activity measurements after 7 and 14 days of continuous exposure to propofol. The model describes transitions from healthy to stressed cells and from stressed cells to death. RESULTS: The model predicted that propofol would affect stressed cells more than healthy cells. Increasing propofol concentration from 10 to 100 μM reduced the mean waiting time for transition to the stressed state by 50%, from 14 to 7 days, whereas the mean duration to cellular death reduced more dramatically from 2.7 days to 6.5 hours. CONCLUSION: The proposed stochastic modeling approach can be used to discriminate between different biological hypotheses regarding the effect of a compound on the transition rates. The effects of different compounds on the transition rate estimates can be quantitatively compared. Data can be extrapolated at late measurement time points to investigate whether costs and time-consuming long-term experiments could possibly be eliminated.

A non-possibility theorem for joint-stability in interindustry models

Joint-stability in interindustry models relates to the mutual simultaneous consistency of the demand-driven and supply-driven models of Leontief and Ghosh, respectively. Previous work has claimed joint-stability to be an acceptable assumption from the empirical viewpoint, provided only small changes in exogenous variables are considered. We show in this note, however, that the issue has deeper theoretical roots and offer an analytical demonstration that shows the impossibility of consistency between demand-driven and supply-driven models.

Entrevistant infants pre-escolars víctimes d’abús sexual i/o maltractament familiar : Eficàcia dels models d’entrevista forense

Entrevistant infants pre-escolars víctimes d’abús sexual i/o maltractament familiar: eficàcia dels models d’entrevista forense Entrevistar infants en edat preescolar que han viscut una situació traumàtica és una tasca complexa que dins l’avaluació psicològica forense necessita d’un protocol perfectament delimitat, clar i temporalitzat. Per això, s’han seleccionat 3 protocols d’entrevista: el Protocol de Menors (PM) de Bull i Birch, el model del National Institute for Children Development (NICHD) de Michel Lamb, a partir del qual es va desenvolupar l’EASI (Evaluación del Abuso Sexual Infantojuvenil) i l’Entrevista Cognitiva (EC) de Fisher i Geiselman. La hipòtesi de partida vol comprovar si els anteriors models permeten obtenir volums informatius diferents en infants preescolars. Conseqüentment, els objectius han estat determinar quin dels models d’entrevista permet obtenir un volum informatiu amb més precisions i menys errors, dissenyar un model d’entrevista propi i consensuar aquest model. En el treball s’afegeixen esquemes pràctics que facilitin l’obertura, desenvolupament i tancament de l’entrevista forense. La metodologia ha reproduït el binomi infant - esdeveniment traumàtic, mitjançant la visualització i l’explicació d’un fet emocionalment significatiu amb facilitat per identificar-se: l’accident en bicicleta d’un infant que cau, es fa mal, sagna i el seu pare el cura. A partir d’aquí, hem entrevistat 135 infants de P3, P4 i P5, mitjançant els 3 models d’entrevista referits, enfrontant-los a una demanda específica: recordar i narrar aquest esdeveniment. S’ha conclòs que el nivell de record correcte, quan s’utilitza un model d’entrevista adequat amb els infants en edat preescolar, oscil•la entre el 70-90%, fet que permet defensar la confiança en els records dels infants. Es constata que el percentatge d’emissions incorrectes dels infants en edat preescolar és mínim, al voltant d’un 5-6%. L’estudi remarca la necessitat d’establir perfectament les regles de l’entrevista i, per últim, en destaca la ineficàcia de les tècniques de memòria de l’entrevista cognitiva en els infants de P3 i P4. En els de P5 es comencen a veure beneficis gràcies a la tècnica de la reinstauració contextual (RC), estant les altres tècniques fora de la comprensió i utilització dels infants d’aquestes edats. Interviewing preschoolers victims of sexual abuse and/or domestic abuse: Effectiveness of forensic interviews models 135 preschool children were interviewed with 3 different interview models in order to remember a significant emotional event. Authors conclude that the correct recall of children ranging from 70-90% and the percentage of error messages is 5-6%. It is necessary to fully establish the rules of the interview. The present research highlights the effectiveness of the cognitive interview techniques in children from P3 and P4. Entrevistando niños preescolares víctimas de abuso sexual y/o maltrato familiar: eficacia de los modelos de entrevista forense Se han entrevistado 135 niños preescolares con 3 modelos de entrevista diferentes para recordar un hecho emocionalmente significativo. Se concluye que el recuerdo correcto de los niños oscila entre el 70-90% y el porcentaje de errores de mensajes es del 5-6%. El estudio remarca la necesidad de establecer perfectamente las reglas de la entrevista y se destaca la ineficacia de las técnicas de la entrevista cognitiva en los niños de P3 y P4.

THE USE OF SIMULATIONS IN EVOLUTIONARY POPULATION GENETICS: APPLICATIONS ON HUMANS, OWLS AND VIRTUAL ORGANISMS

Summary (in English) Computer simulations provide a practical way to address scientific questions that would be otherwise intractable. In evolutionary biology, and in population genetics in particular, the investigation of evolutionary processes frequently involves the implementation of complex models, making simulations a particularly valuable tool in the area. In this thesis work, I explored three questions involving the geographical range expansion of populations, taking advantage of spatially explicit simulations coupled with approximate Bayesian computation. First, the neutral evolutionary history of the human spread around the world was investigated, leading to a surprisingly simple model: A straightforward diffusion process of migrations from east Africa throughout a world map with homogeneous landmasses replicated to very large extent the complex patterns observed in real human populations, suggesting a more continuous (as opposed to structured) view of the distribution of modern human genetic diversity, which may play a better role as a base model for further studies. Second, the postglacial evolution of the European barn owl, with the formation of a remarkable coat-color cline, was inspected with two rounds of simulations: (i) determine the demographic background history and (ii) test the probability of a phenotypic cline, like the one observed in the natural populations, to appear without natural selection. We verified that the modern barn owl population originated from a single Iberian refugium and that they formed their color cline, not due to neutral evolution, but with the necessary participation of selection. The third and last part of this thesis refers to a simulation-only study inspired by the barn owl case above. In this chapter, we showed that selection is, indeed, effective during range expansions and that it leaves a distinguished signature, which can then be used to detect and measure natural selection in range-expanding populations. Résumé (en français) Les simulations fournissent un moyen pratique pour répondre à des questions scientifiques qui seraient inabordable autrement. En génétique des populations, l'étude des processus évolutifs implique souvent la mise en oeuvre de modèles complexes, et les simulations sont un outil particulièrement précieux dans ce domaine. Dans cette thèse, j'ai exploré trois questions en utilisant des simulations spatialement explicites dans un cadre de calculs Bayésiens approximés (approximate Bayesian computation : ABC). Tout d'abord, l'histoire de la colonisation humaine mondiale et de l'évolution de parties neutres du génome a été étudiée grâce à un modèle étonnement simple. Un processus de diffusion des migrants de l'Afrique orientale à travers un monde avec des masses terrestres homogènes a reproduit, dans une très large mesure, les signatures génétiques complexes observées dans les populations humaines réelles. Un tel modèle continu (opposé à un modèle structuré en populations) pourrait être très utile comme modèle de base dans l'étude de génétique humaine à l'avenir. Deuxièmement, l'évolution postglaciaire d'un gradient de couleur chez l'Effraie des clocher (Tyto alba) Européenne, a été examiné avec deux séries de simulations pour : (i) déterminer l'histoire démographique de base et (ii) tester la probabilité qu'un gradient phénotypique, tel qu'observé dans les populations naturelles puisse apparaître sans sélection naturelle. Nous avons montré que la population actuelle des chouettes est sortie d'un unique refuge ibérique et que le gradient de couleur ne peux pas s'être formé de manière neutre (sans l'action de la sélection naturelle). La troisième partie de cette thèse se réfère à une étude par simulations inspirée par l'étude de l'Effraie. Dans ce dernier chapitre, nous avons montré que la sélection est, en effet, aussi efficace dans les cas d'expansion d'aire de distribution et qu'elle laisse une signature unique, qui peut être utilisée pour la détecter et estimer sa force.