The Impact of Aging Drivers and Vehicles on the Injury Severity of Crash Victims.

Against a general trend of increasing driver longevity, the injuries suffered by vehicle occupants in Spanish road traffic crashes are analyzed by the level of severity of their bodily injuries (BI). Generalized linear mixed models are applied to model the proportion of non-serious, serious, and fatal victims. The dependence between vehicles involved in the same crash is captured by including random effects. The effect of driver age and vehicle age and their interaction on the proportion of injured victims is analyzed. We find a nonlinear relationship between driver age and BI severity, with young and older drivers constituting the riskiest groups. In contrast, the expected severity of the crash increases linearly up to a vehicle age of 18 and remains constant thereafter at the highest level of BI severity. No interaction between the two variables is found. These results are especially relevant for countries such as Spain with increasing driver longevity and an aging car fleet.

An ensemble learning strategy for panel time series forecasting of excess mortality during the COVID-19 pandemic.

Quantifying and analyzing excess mortality in crises such as the ongoing COVID-19 pandemic is crucial for policymakers. Traditional measures fail to take into account differences in the level, long-term secular trends, and seasonal patterns in all-cause mortality across countries and regions. This paper develops and empirically investigates the forecasting performance of a novel, flexible and dynamic ensemble learning with a model selection strategy (DELMS) for the seasonal time series forecasting of monthly respiratory disease death data across a pool of 61 heterogeneous countries. The strategy is based on a Bayesian model averaging (BMA) of heterogeneous time series methods involving both the selection of the subset of best forecasters (model confidence set), the identification of the best holdout period for each contributed model, and the determination of optimal weights using out-of-sample predictive accuracy. A model selection strategy is also developed to remove the outlier models and to combine the models with reasonable accuracy in the ensemble. The empirical outcomes of this large set of experiments show that the accuracy of the BMA approach is significantly improved with DELMS when selecting a flexible and dynamic holdout period and removing the outlier models. Additionally, the forecasts of respiratory disease deaths for each country are highly accurate and exhibit a high correlation (94%) with COVID-19 deaths in 2020.

Does longevity impact the severity of traffic crashes? A comparative study of young-older and old-older drivers.

INTRODUCTION: This article analyzes the effect of driver's age in crash severity with a particular focus on those over the age of 65. The greater frequency and longevity of older drivers around the world suggests the need to introduce a possible segmentation within this group at risk, thus eliminating the generic interval of 65 and over as applied today in road safety data and in the automobile insurance sector. METHOD: We investigate differences in the severity of traffic crashes among two subgroups of older drivers -young-older (65-75) and old-older (75+), and findings are compared with the age interval of drivers under 65. Here, we draw on data for 2016 provided by Spanish Traffic Authority. Parametric and semi-parametric regression models are applied. RESULTS: We identified the factors related to the crash, vehicle, and driver that have a significant impact on the probability of the crash being slight, serious, or fatal for the different age groups. CONCLUSIONS: We found that crash severity and the expected costs of crashes significantly increase when the driver is over the age of 75. Practical Applications: Our results have obvious implications for regulators responsible for road safety policies - most specifically as they consider there should be specific driver licensing requirements and driving training for elderly - and for the automobile insurance industry, which to date has not examined the impact that the longevity of drivers is likely to have on their balance sheets.

The Use of Telematics Devices to Improve Automobile Insurance Rates.

Most automobile insurance databases contain a large number of policyholders with zero claims. This high frequency of zeros may reflect the fact that some insureds make little use of their vehicle, or that they do not wish to make a claim for small accidents in order to avoid an increase in their premium, but it might also be because of good driving. We analyze information on exposure to risk and driving habits using telematics data from a pay-as-you-drive sample of insureds. We include distance traveled per year as part of an offset in a zero-inflated Poisson model to predict the excess of zeros. We show the existence of a learning effect for large values of distance traveled, so that longer driving should result in higher premiums, but there should be a discount for drivers who accumulate longer distances over time due to the increased proportion of zero claims. We confirm that speed limit violations and driving in urban areas increase the expected number of accident claims. We discuss how telematics information can be used to design better insurance and to improve traffic safety.

Copula-based regression modeling of bivariate severity of temporary disability and permanent motor injuries.

The analysis of factors influencing the severity of the personal injuries suffered by victims of motor accidents is an issue of major interest. Yet, most of the extant literature has tended to address this question by focusing on either the severity of temporary disability or the severity of permanent injury. In this paper, a bivariate copula-based regression model for temporary disability and permanent injury severities is introduced for the joint analysis of the relationship with the set of factors that might influence both categories of injury. Using a motor insurance database with 21,361 observations, the copula-based regression model is shown to give a better performance than that of a model based on the assumption of independence. The inclusion of the dependence structure in the analysis has a higher impact on the variance estimates of the injury severities than it does on the point estimates. By taking into account the dependence between temporary and permanent severities a more extensive factor analysis can be conducted. We illustrate that the conditional distribution functions of injury severities may be estimated, thus, providing decision makers with valuable information.

Time and distance to first accident and driving patterns of young drivers with pay-as-you-drive insurance.

We conducted a study of approximately 16,000 drivers under the age of 30 that had purchased a pay-as-you-drive insurance policy, where their risk of being involved in a crash was analyzed from vehicle tracking data using a global positioning system. The comparison of novice vs. experienced young drivers shows that vehicle usage differs significantly between these groups and that the time to the first crash is shorter for those drivers with less experience. Driving at night and a higher proportion of speed limit violations reduces the time to the first crash for both novice and experienced young drivers, while urban driving reduces the distance traveled to the first crash for both groups. Gender differences are also observed in relation to the influence of driving patterns on the risk of accident. Nighttime driving reduces the time to the first accident in the case of women, but not for men. The risk of an accident increases with excessive speed, but the effect of speed is significantly higher for men than it is for women among the more experienced drivers.

Impact of road traffic injuries on disability rates and long-term care costs in Spain.

Road traffic injuries are one of the leading causes of increasing disability-adjusted life expectancy. We analyze long-term care needs associated with motor vehicle crash-related disability in Spain and conclude that needs attributable traffic injuries are most prevalent during victims' mid-life years, they create a significant burden for both families and society as a whole given that public welfare programmes supporting these victims need to be maintained over a long time frame. High socio-economic costs of road traffic accidents (in Spain 0.04% of the GDP in 2008) are clearly indicative of the need for governments and policymakers to strengthen road accident preventive measures.

The impact of traffic violations on the estimated cost of traffic accidents with victims.

We analyse accidents with victims and calculate the influence of traffic violations on the probability of having a serious or fatal accident, compared to a slight accident. Traffic violations related to speed limitations, administrative infringements or faults related to the driver are considered. Data were obtained from all available reports on accidents with victims that occurred in Spain from 2003 to 2005. A multinomial logistic regression model is specified to find the probability that an accident with victims is slight, serious or fatal, given the presence/absence of thirty different types of traffic violations. The average cost per victim and the average number of victims per accident are then used to find the estimated cost of an accident with victims, given the information on the traffic violations incurred. This demonstrates which combinations of traffic violations lead to higher estimated average costs, compared to cases in which no traffic violation occurred. We conclude with some recommendations on the severity of penalties, and suggest that regulators penalize the occurrences of some specific combinations of traffic violations more rigorously.

[The importance of the sample design effect]. / La importancia del efecto del diseño.

INTRODUCTION: Sample selection through a complex design influences the subsequent statistical analysis. MATERIAL AND METHOD: The different means of sample selection may result in bias and greater variance of estimators; simple randomized sampling is the reference design. Diverse examples are provided, illustrating how the various sampling strategies can result in bias and increase variance. RESULTS: The inclusion of different weighting techniques reduces bias. Evaluation of the effect of design enables measurement of the degree of variance distortion due to the sampling design used and therefore provides a direct evaluation of the alteration in the confidence intervals estimated when the sampling design deviates from simple randomized sampling. CONCLUSIONS: We recommend measurement of the effect of the design on analysis of the data obtained by sampling and inclusion of weighting techniques in statistical analyses.

La importancia del efecto del diseño / The importance of the sample design effect

INTRODUCCIÓN: La selección de una muestra mediante un diseño complejo tiene consecuencias en el análisis estadístico posterior. MATERIAL Y MÉTODO: Las diferentes formas de selección muestral pueden acarrear la aparición de sesgo y de una mayor varianza de los estimadores y es el muestreo aleatorio simple el diseño de referencia.Se proporcionan diversos ejemplos ilustrativos en los que diferentes estrategias de muestreo conducen a situaciones de aparición de sesgo y de aumento de la varianza. RESULTADOS: La inclusión de ponderaciones conduce a la eliminación de sesgo. La evaluación del efecto del diseño permite medir el grado de distorsión que sufren las varianzas debido al diseño muestral empleado y, por lo tanto, proporciona una valoración directa de la alteración que sufren los intervalos de confianza estimados cuando el diseño muestral se aparta del caso aleatorio simple. CONCLUSIONES: Se recomienda emplear una medida del efecto del diseño en análisis de datos obtenidos por muestreo e incluir ponderaciones en los análisis estadísticos (AU)