Claim Amount Forecasting and Pricing of Automobile Insurance Based on the BP Neural Network

The BP neural network model is a hot issue in recent academic research, and it has been successfully applied to many other fields, but few researchers apply the BP neural network model to the field of automobile insurance. The main method that has been used in the prediction of the total claim amount in automobile insurance is the generalized linear model, where the BP neural network model could provide a different approach to estimate the total claim loss. This paper uses a genetic algorithm to optimize the structure of the BP neural network at first, and the calculation speed is significantly improved. At the same time, by considering the overfitting problem, an early stop method is introduced to avoid the overfitting problem. In the model, a three-layer BP neural network model, which includes the input layer, hidden layer, and output layer, is trained. With consideration of various factors, a total claim amount prediction model is established, and the trained BP neural network model is used to predict the total claim amount of automobile insurance based on the data of the training set. The results show that the accuracy of the prediction by using the BP neural network model to both the data of Shandong Province and to the data of six cities is over 95%. Then, the predicted total claim amount is used to calculate premiums for five cities in Shandong Province according to credibility theory. The results show that the average premium of the five cities is slightly higher than the actual claim amount of the city. The combination of BP neural network and credibility theory can perform accurate claim amount estimation and pricing for automobile insurance, which can effectively improve the current situation of the automobile insurance business and promote the development of insurance industry.

On the total claim amount for marked Poisson cluster models

We study the asymptotic distribution of the total claim amount for marked Poisson cluster models. The marks determine the size and other characteristics of the individual claims and potentially influence the arrival rate of future claims. We find sufficient conditions under which the total claim amount satisfies the central limit theorem or, alternatively, tends in distribution to an infinite-variance stable random variable. We discuss several Poisson cluster models in detail, paying special attention to the marked Hawkes process as our key example.

The Copula-Based Total Claim Amount Regression Model with an Unobserved Risk Factor

Nowadays a common practice of any insurance company is ratemaking, which is defined as the process of classification of the mass risk portfolio into risk groups where the same premium corresponds to each risk. As generalised linear models are usually applied, the process requires the independence between the average value of claims and the number of claims. However, in literature this assumption is called into question. The interest of this paper is to propose the copula-based total claim amount model taking into account an unobservable risk factor in the claim frequency model. This factor, called also as unobserved heterogeneity, is treated as a random variable influencing the number of claims. The goal is to estimate the expected value of the product of two random variables: the average value of claims and the number of claims for a single risk assuming the dependence between the average value of claims and the number of claims for a single risk and the dependence between the number of claims for a single risk and the unobservable risk factor. We give details of the theoretical aspects of the model as well as the empirical example. To acquaint the reader with the model operation, every step of the process of the expected value estimation in described and the R code is available for download, see http://web.ue.katowice.pl/woali/.

COMPARISON OF APPROXIMATIONS FOR COMPOUND POISSON PROCESSES

In this paper, we compare the error in several approximation methods for the cumulative aggregate claim distribution customarily used in the collective model of insurance theory. In this model, it is usually supposed that a portfolio is at risk for a time period of length t. The occurrences of the claims are governed by a Poisson process of intensity μ so that the number of claims in [0,t] is a Poisson random variable with parameter λ = μ t. Each single claim is an independent replication of the random variable X, representing the claim severity. The aggregate claim or total claim amount process in [0,t] is represented by the random sum of N independent replications of X, whose cumulative distribution function (cdf) is the object of study. Due to its computational complexity, several approximation methods for this cdf have been proposed. In this paper, we consider 15 approximations put forward in the literature that only use information on the lower order moments of the involved distributions. For each approximation, we consider the difference between the true distribution and the approximating one and we propose to use expansions of this difference related to Edgeworth series to measure their accuracy as λ = μ t diverges to infinity. Using these expansions, several statements concerning the quality of approximations for the distribution of the aggregate claim process can find theoretical support. Other statements can be disproved on the same grounds. Finally, we investigate numerically the accuracy of the proposed formulas.

On Premium Estimation Using the C&RT/Poisson Model and Its Extensions

Premium estimation is a key concept in insurance mathematics. Estimation of the mean andvariance of a total claim amount of a portfolio can be considered as necessary prerequisites for this. Inturn, dividing the portfolio into homogeneous subportfolios can be considered as a rst step towards ndingthose estimates. We consider the problem of estimating the claim intensity and propose a regressiontrees based approach for clustering the portfolio into homogeneous subportfolios in a situation where thedurations of the policies dier and overdispersion is present. Several other generalizations are discussed.A case study involving Estonian casco insurance is included.

Estimation of the probable maximum loss based on extreme value theory for stock returns

Probable maximum loss is a measure coming from the insurance market, where is applied to insurance portfolio analysis. This correspond to the 20-80 rule, which states that 20% of the individual claims are responsible for more than 80% of the total claim amount in a well defined portfolio. The main aim of the presented paper is estimation of the probable maximum loss for stock returns which are treated as portfolios of securities. It turns out that probable maximum loss is a useful tool for risk analysis or/and diagnostic purposes at capital markets, but we have to be aware of its drawbacks.