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.

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.

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.

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

Clinical consequences of switching from olanzapine to risperidone and vice versa in outpatients with schizophrenia: 36-month results from the worldwide schizophrenia outpatients health outcomes (W-SOHO) study

BACKGROUND: With many atypical antipsychotics now available in the market, it has become a common clinical practice to switch between atypical agents as a means of achieving the best clinical outcomes. This study aimed to examine the impact of switching from olanzapine to risperidone and vice versa on clinical status and tolerability outcomes in outpatients with schizophrenia in a naturalistic setting. METHODS: W-SOHO was a 3-year observational study that involved over 17,000 outpatients with schizophrenia from 37 countries worldwide. The present post hoc study focused on the subgroup of patients who started taking olanzapine at baseline and subsequently made the first switch to risperidone (n=162) and vice versa (n=136). Clinical status was assessed at the visit when the first switch was made (i.e. before switching) and after switching. Logistic regression models examined the impact of medication switch on tolerability outcomes, and linear regression models assessed the association between medication switch and change in the Clinical Global Impression-Schizophrenia (CGI-SCH) overall score or change in weight. In addition, Kaplan-Meier survival curves and Cox-proportional hazards models were used to analyze the time to medication switch as well as time to relapse (symptom worsening as assessed by the CGI-SCH scale or hospitalization). RESULTS: 48% and 39% of patients switching to olanzapine and risperidone, respectively, remained on the medication without further switches (p=0.019). Patients switching to olanzapine were significantly less likely to experience relapse (hazard ratio: 3.43, 95% CI: 1.43, 8.26), extrapyramidal symptoms (odds ratio [OR]: 4.02, 95% CI: 1.49, 10.89) and amenorrhea/galactorrhea (OR: 8.99, 95% CI: 2.30, 35.13). No significant difference in weight change was, however, found between the two groups. While the CGI-SCH overall score improved in both groups after switching, there was a significantly greater change in those who switched to olanzapine (difference of 0.29 points, p=0.013). CONCLUSION: Our study showed that patients who switched from risperidone to olanzapine were likely to experience a more favorable treatment course than those who switched from olanzapine to risperidone. Given the nature of observational study design and small sample size, additional studies are warranted.

Constant coefficient tests for random coefficient regression

Random coefficient regression models have been applied in differentfields and they constitute a unifying setup for many statisticalproblems. The nonparametric study of this model started with Beranand Hall (1992) and it has become a fruitful framework. In thispaper we propose and study statistics for testing a basic hypothesisconcerning this model: the constancy of coefficients. The asymptoticbehavior of the statistics is investigated and bootstrapapproximations are used in order to determine the critical values ofthe test statistics. A simulation study illustrates the performanceof the proposals.

Assessing the structural change of strategic mobility: determinants under hypercompetitive environments

The primary purpose of this exploratory empirical study is to examine the structural stability of a limited number of alternative explanatory factors of strategic change. On the basis of theoretical arguments and prior empirical evidence from two traditional perspectives, we propose an original empirical framework to analyse whether these potential explanatory factors have remained stable over time in a highly turbulent environment. This original question is explored in a particular setting: the population of Spanish private banks. The firms of this industry have experienced a high level of strategic mobility as a consequence of fundamental changes undergone in their environmental conditions over the last two decades (mainly changes related to the new banking and financial regulation process). Our results consistently support that the effect of most explanatory factors of strategic mobility considered did not remain stable over the whole period of analysis. From this point of view, the study sheds new light on major debates and dilemmas in the field of strategy regarding why firms change their competitive patterns over time and, hence, to what extent the "contextdependency" of alternative views of strategic change as their relative validation can vary over time for a given population. Methodologically, this research makes two major contributions to the study of potential determinants of strategic change. First, the definition and measurement of strategic change employing a new grouping method, the Model-based Cluster Method or MCLUST. Second, in order to asses the possible effect of determinants of strategic mobility we have controlled the non-observable heterogeneity using logistic regression models for panel data.

Locations and Relocations:Modelling, Determinants, and Interrelations

Empirical studies on industrial location do not typically distinguish between new and relocated establishments. This paper addresses this shortcoming using data on the frequency of these events in municipalities of the same economic-administrative region. This enables us to test not only for differences in their determinants but also for interrelations between start-ups and relocations. Estimates from count regression models for cross-section and panel data show that, although partial effects differ, common patterns arise in “institutional” and “neoclassical” explanatory factors. Also, start-ups and relocations are positive but asymmetrically related. JEL classification: C25, R30, R10. Keywords: cities, count data models, industrial location

Does grade retention affect achievement? Some evidence from PISA

Grade retention practices are at the forefront of the educational debate. In this paper, we use PISA 2009 data for Spain to measure the effect of grade retention on students’ achievement. One important problem when analyzing this question is that school outcomes and the propensity to repeat a grade are likely to be determined simultaneously. We address this problem by estimating a Switching Regression Model. We …find that grade retention has a negative impact on educational outcomes, but we confi…rm the importance of endogenous selection, which makes observed differences between repeaters and non-repeaters appear 14.6% lower than they actually are. The effect on PISA scores of repeating is much smaller (-10% of non-repeaters’ average) than the counterfactual reduction that non-repeaters would suffer had they been retained as repeaters (-24% of their average). Furthermore, those who repeated a grade during primary education suffered more than those who repeated a grade of secondary school, although the effect of repeating at both times is, as expected, much larger.

Growth in a Cross-Section of Cities: Location, Increasing Returns or Random Growth?

This article analyzes empirically the main existing theories on income and population city growth: increasing returns to scale, locational fundamentals and random growth. To do this we implement a threshold nonlinearity test that extends standard linear growth regression models to a dataset on urban, climatological and macroeconomic variables on 1,175 U.S. cities. Our analysis reveals the existence of increasing returns when per-capita income levels are beyond $19; 264. Despite this, income growth is mostly explained by social and locational fundamentals. Population growth also exhibits two distinct equilibria determined by a threshold value of 116,300 inhabitants beyond which city population grows at a higher rate. Income and population growth do not go hand in hand, implying an optimal level of population beyond which income growth stagnates or deteriorates

Social network measures for "nosduocentered" networks, their predictive power on performance

Our purpose in this article is to define a network structure which is based on two egos instead of the egocentered (one ego) or the complete network (n egos). We describe the characteristics and properties for this kind of network which we call “nosduocentered network”, comparing it with complete and egocentered networks. The key point for this kind of network is that relations exist between the two main egos and all alters, but relations among others are not observed. After that, we use new social network measures adapted to the nosduocentered network, some of which are based on measures for complete networks such as degree, betweenness, closeness centrality or density, while some others are tailormade for nosduocentered networks. We specify three regression models to predict research performance of PhD students based on these social network measures for different networks such as advice, collaboration, emotional support and trust. Data used are from Slovenian PhD students and their s

Networks of PhD Students and Academic Performance: A Comparison across Countries

In this article we compare regression models obtained to predict PhD students’ academic performance in the universities of Girona (Spain) and Slovenia. Explanatory variables are characteristics of PhD student’s research group understood as an egocentered social network, background and attitudinal characteristics of the PhD students and some characteristics of the supervisors. Academic performance was measured by the weighted number of publications. Two web questionnaires were designed, one for PhD students and one for their supervisors and other research group members. Most of the variables were easily comparable across universities due to the careful translation procedure and pre-tests. When direct comparison was notpossible we created comparable indicators. We used a regression model in which the country was introduced as a dummy coded variable including all possible interaction effects. The optimal transformations of the main and interaction variables are discussed. Some differences between Slovenian and Girona universities emerge. Some variables like supervisor’s performance and motivation for autonomy prior to starting the PhD have the same positive effect on the PhD student’s performance in both countries. On the other hand, variables like too close supervision by the supervisor and having children have a negative influence in both countries. However, we find differences between countries when we observe the motivation for research prior to starting the PhD which increases performance in Slovenia but not in Girona. As regards network variables, frequency of supervisor advice increases performance in Slovenia and decreases it in Girona. The negative effect in Girona could be explained by the fact that additional contacts of the PhD student with his/her supervisor might indicate a higher workload in addition to or instead of a better advice about the dissertation. The number of external student’s advice relationships and social support mean contact intensity are not significant in Girona, but they have a negative effect in Slovenia. We might explain the negative effect of external advice relationships in Slovenia by saying that a lot of external advice may actually result from a lack of the more relevant internal advice

Internal test sets studies in a group of antimalarials

Topological indices have been applied to build QSAR models for a set of 20 antimalarial cyclic peroxy cetals. In order to evaluate the reliability of the proposed linear models leave-n-out and Internal Test Sets (ITS) approaches have been considered. The proposed procedure resulted in a robust and consensued prediction equation and here it is shown why it is superior to the employed standard cross-validation algorithms involving multilinear regression models

Health inequalities among the elderly in western Europe

Background: This paper analyses gender inequalities in health status and in social determinants of health among the elderly in Western Europe. Methods: Data came from the first wave of the “Survey of Health, Ageing and Retirement in Europe” (SHARE, 2004). For the purposes of this study a subsample of community-residing people aged 65-85 years with no paid work was selected (4218 men and 5007 women). Multiple logistic regression models separated by sex and adjusted for age and country were fitted. Results: Women were more likely to report poor health status, limitations in mobility and poor mental health. Whereas in both sexes educational attainment was associated with the three health indicators, household income was only related to poor self-rated health among women. The relationship between living arrangements and health differed by gender and was primarily associated with poor mental health. In both sexes, not living with the partner but living with other people and being the household head was related to poor mental health status (aOR=2.14; 95% CI=1.11-4.14 for men and aOR=1.75; 95% CI=1.12-2.72 for women). Additionally, women living with their partner and other(s) and those living alone were more likely to report poor mental health status (aOR=1.67; 95% CI=1.17-2.41 and aOR=1.58; 95% CI=1.26-1.97, respectively). Conclusions: Health inequalities persist among the elderly. Women have poorer health status than men and in both sexes the risk of poor health status increases among those with low educational attainment. Living arrangements are primarily associated with poor mental health status with patterns that differ by gender.

Gender inequality in health among elderly people in a combined framework of socioeconomic position, family characteristics and social support

This study analyses gender inequalities in health among elderly people in Catalonia (Spain) by adopting a conceptual framework that globally considers three dimensions of health determinants : socio-economic position, family characteristics and social support. Data came from the 2006 Catalonian Health Survey. For the purposes of this study a sub-sample of people aged 65–85 years with no paid job was selected (1,113 men and 1,484 women). The health outcomes analysed were self-perceived health status, poor mental health status and long-standing limiting illness. Multiple logistic regression models separated by sex were fitted and a hierarchical model was fitted in three steps. Health status among elderly women was poorer than among the men for the three outcomes analysed. Whereas living with disabled people was positively related to the three health outcomes and confidant social support was negatively associated with all of them in both sexes, there were gender differences in other social determinants of health. Our results emphasise the importance of using an integrated approach for the analysis of health inequalities among elderly people, simultaneously considering socio-economic position, family characteristics and social support, as well as different health indicators, in order fully to understand the social determinants of the health status of older men and women.