IMPROVING AUTOMOBILE INSURANCE CLAIMS FREQUENCY PREDICTION WITH TELEMATICS CAR DRIVING DATA

Novel navigation applications provide a driving behavior score for each finished trip to promote safe driving, which is mainly based on experts’ domain knowledge. In this paper, with automobile insurance claims data and associated telematics car driving data, we propose a supervised driving risk scoring neural network model. This one-dimensional convolutional neural network takes time series of individual car driving trips as input and returns a risk score in the unit range of (0,1). By incorporating credibility average risk score of each driver, the classical Poisson generalized linear model for automobile insurance claims frequency prediction can be improved significantly. Hence, compared with non-telematics-based insurers, telematics-based insurers can discover more heterogeneity in their portfolio and attract safer drivers with premiums discounts.

Classification Ratemaking Using Decision Tree in the Insurance Market of Bosnia and Herzegovina

This paper investigates the impact of risk classification on life insurance ratemaking with particular reference to Bosnia and Herzegovina (BiH). The research is based on a sample of over eighteen thousand insurance policies for passenger vehicles collected over the period 2015-2020. In our empirical investigation we develop a standard risk model based on the application of Poisson Generalized linear models (GLM) for claims frequency estimate and Gamma GLM for claim severity estimate. The analysis reveals that GLM does not provide a reliable parameter estimates for Multi-level factor (MLF) categorical predictors. Although GLM is widely used method to deter insurance premiums, improvements of GLM by using the data mining methods identified in this paper may solve practical challenges for the risk models. The popularity of applying data mining methods in the actuarial community has been growing in recent years due to its efficiency and precision. These models are recommended to be considered in BiH and South East European region in general.

Nutrition Claims Frequency and Compliance in a Food Sample of the Spanish Market: The BADALI Study

Nutrition claims (NCs) have been shown to affect customers’ perceptions and behaviour. In Europe, they are regulated by Regulation (EC) No 1924/2006. The aim of this work was to analyse the prevalence and compliance of NCs according to this regulation in Spain. For this purpose, we used the BADALI database, which included 3197 foods present in the Spanish market. Our results show that 36.1% of all foods carried NCs, at a rate of 3.3 NCs/food. The prevalence was very heterogeneous among food groups. Nuts and seeds, legumes and non-alcoholic beverages were the groups with the highest prevalence. Micronutrients, fat, fibre and sugars were the nutrients most referred to in NCs. Overall, the compliance was low, with 49.2% NCs correct. Fibre and proteins were the nutrients with most correct NCs. Vegetables and non-alcoholic beverages were the food groups with the highest proportion of correct NCs. The main reason for incorrect NCs was because the amount of the nutrient was not stated in the label. The results of our study reveal that the aim of the European Commission to ensure a high level of protection for consumers regarding NCs has not been fulfilled. Therefore, we consider it crucial that European institutions invest in guaranteeing regulation compliance.

The impact of telematics on the insurability of risks

Purpose The purpose of this paper is to analyze the use of telematics in insurance and its consequences for the insurability of risks. Empirical results on monitoring policyholders or insured objects and its consequences for asymmetric information, as well as claims frequency and severity are discussed. Furthermore, potential future research questions that arise from the use of telematics in risk management and insurance are outlined. Design/methodology/approach The paper systematically reviews existing studies and then investigates the consequences of telematics using Berliner’s insurability criteria. The results are based on 52 academic studies and industry papers published from 2000 to 2019. Findings The findings emphasize the effects of new information on information asymmetry and risk pooling, the implications of new technologies on loss frequency and severity, legal restrictions and ethical consequences of the use of telematics in the insurance field. Problems with the insurability impede the market development of innovations such as telematics tariffs. Originality/value Despite its increasing relevance for businesses at present, research on telematics in insurance is limited. Some papers can be found in the IT domain, but relatively little research has been done in the business and economics literature. The authors illustrate where the research stands currently and outline directions for future research.

The Over-Searching Accidents Causative Factors in Ghana: The Role of Policyholders Education Levels

Available statistics indicates that about 90% of all claims or accident in Ghana is caused by human behavior. Therefore, policyholders’ errors are categorized depending on the severity and extend of casualties caused as a result of misinterpretation of road traffic control devices based on their education levels. Hence, in order to ascertain all the possible causes within the human element to reduce the increasing trend of yearly claims, this study report on the influence of education levels on accident/claims frequency and severity drawing upon a purposive sample of 203 policyholders who have experienced at least one accident in a year using structural equation modeling (SEM). The findings from our regression weights gave enough evidence to reject most of our hypotheses with few ones being supported. This study provides enough evidence that education generally to perspective policyholders influence accidents/claims occurrence. However, in terms of education levels of policyholders, we did not have enough evidence in support of any of these levels either causing or reducing claims/accident frequency. Besides accident/claim frequency, we extended our regression analysis on claim severity and also included some well know auto insurance rating factors to ascertain their impacts on accident frequency. Consequently, it was revealed that most of the severe claims or accidents that results into deaths and serious injuries on yearly basis are caused by policyholders or drivers with medium level of education in Ghana with its frequency driving mostly by rating factors such as the vehicle’s age, cubic capacity, mileage, etc.

A plan of capital injections based on the claims frequency

In this work, we propose a capital injection strategy which is periodically implemented based on the number of claims in the classical Poisson risk model. Especially, capital injection decisions are made at a predetermined accumulated number of claim instants, if the surplus is lower than a minimum required level. There appears to be a similar problem found in reliability theory such that preventive maintenance policies are performed at certain shock numbers. Assuming a combination of exponentials for the claim severities, we first derive an explicit expression for the discounted density of the surplus level after a certain number of claims if ruin has not yet occurred. Utilising this result, we study the expected total discounted capital injection until the first ruin time. To solve the differential equation associated with this quantity, we analyse an extended Lundberg’s fundamental equation. Similarly, an expression for the Laplace transform of the time to ruin is also explicitly found. Finally, we illustrate the applicability of the present capital injection strategy and methodologies through various numerical examples. In particular, for exponential claim severities, some optimal capital injection strategy which minimises the expected capital spending per unit time is numerically studied.

Bonus–Malus systems with Weibull distributed claim severities

One of the pricing strategies for Bonus–Malus (BM) systems relies on the decomposition of the claims’ randomness into one part accounting for claims’ frequency and the other part for claims’ severity. By mixing an exponential with a Lévy distribution, we focus on modelling the claim severity component as a Weibull distribution. For a Negative Binomial number of claims, we employ the Bayesian approach to derive the BM premiums for Weibull severities. We then conclude by comparing our explicit formulas and numerical results with those for Pareto severities that were introduced by Frangos & Vrontos.

Experience Rating, Work Injuries and Benefit Costs

In 1986, British Columbia’s Workers’ Compensation Board introduced an experience rating program that provided a modest financial incentive for employers to reduce the costs of claims. Using a comprehensive panel data set, we find that claims frequency for health care only and short-term disability claims was reduced following the introduction of experience rating. The introduction of the program did not affect costs for most claim types, except for health care only claims.