A DEA analysis of risk,cost, and revenues in insurance

Insurance companies have to take risk and cost into account when pricing car insurance policies that cover the risk of private use of cars. In this paper we use data from 80?000 car insurance policies in order to assess, once risk and cost have been taken into account, the combinations of risk that generate the highest returns for the company under existing pricing practices. We use data envelopment analysis (DEA) and frame the study within an analysis of experiments context. The results of DEA are interpreted in a multivariate statistical analysis context using factor analysis, and property fitting techniques. The impact of risk factors in the efficiency is explored by means of regression analysis with dummy variables. There are consequences for the pricing policy of the company.     

Modeling the number of insureds’ cars using queuing theory

In this paper, we propose to model the number of insured cars per household. We use queuing theory to construct a new model that needs 4 different parameters: one that describes the rate of addition of new cars on the insurance contract, a second one that models the rate of removal of insured vehicles, a third parameter that models the cancellation rate of the insurance policy, and finally a parameter that describes the rate of renewal. Statistical inference techniques allow us to estimate each parameter of the model, even in the case where there is censorship of data. We also propose to generalize this new queuing process by adding some explanatory variables into each parameter of the model. This allows us to determine which policyholder’s profiles are more likely to add or remove vehicles from their insurance policy, to cancel their contract or to renew annually. The estimated parameters help us to analyze the insurance portfolio in detail because the queuing theory model allows us to compute various kinds of useful statistics for insurers, such as the expected number of cars insured or the customer lifetime value that calculates the discounted future profits of an insured. Using car insurance data, a numerical illustration based on a portfolio from a Canadian insurance company is included to support this discussion.     

Calculation of Property Insurance Premiums from Statistical Data

A method is proposed for calculation of premiums for a class of insurance risks in an inhomogeneous portfolio. The premiums are calculated using statistical data collected over time for different risk classes. The data represent various contract amounts and various levels of insurance claims. The calculations are carried out by the maximum likelihood method.     

变异系数的区间估计

在机动车辆索赔次数模型的研究中,考虑分布的离散程度是必要的,它有助于选择正确的统计模型。对于给定的样本容量为T的一个样本Ni,i=1,…,T,本文用非参数方法给出了KarlPear son变异系数的区间估计。最后将估计方法应用于我国保险公司的一组实际数据,得到令人满意的结论。     

On the effectiveness of scenario generation techniques in single-period portfolio optimization

In single-period portfolio selection problems the expected value of both the risk measure and the portfolio return have to be estimated. Historical data realizations, used as equally probable scenarios, are frequently used to this aim. Several other parametric and non-parametric methods can be applied. When dealing with scenario generation techniques practitioners are mainly concerned on how reliable and effective such methods are when embedded into portfolio selection models. In this paper we survey different techniques to generate scenarios for the rates of return. We also compare the techniques by providing in-sample and out-of-sample analysis of the portfolios obtained by using these techniques to generate the rates of return. Evidence on the computational burden required by the different techniques is also provided. As reference model we use the Worst Conditional Expectation model with transaction costs. Extensive computational results based on different historical data sets from London Stock Exchange Market (FTSE) are presented and some interesting financial conclusions are drawn.     

GAM在汽车保险定价中的应用研究

汽车保险定价的基础在于风险分析,车辆、驾驶人以及行车环境等因素构成汽车保险定价所倚赖的一个风险系统.本文引入广义加法模型(GAM).将非参数平滑方法应用于到GAM中,结合贝叶斯理论(Bayes)和马尔可夫蒙特卡罗(MCMC)方法得到参数估计,构建汽车保险定价模型,并以国外某保险数据为样本进行实证分析,得到了较好的效果.     

Geoadditive regression for analyzing small-scale geographical variability in car insurance

The empirical part of this article is based on a study on car insurance data to explore how global and local geographical effects on frequency and size of claims can be assessed with appropriate statistical methodology. Because these spatial effects have to be modeled and estimated simultaneously with linear and possibly nonlinear effects of further covariates such as age of policy holder, age of car or bonus-malus score, generalized linear models cannot be applied. Also, compared to previous analyses, the geographical information is given by the exact location of the residence of policy holders. Therefore, we employ a new class of geoadditive models, where the spatial component is modeled based on stationary Gaussian random fields, common in geostatistics (Kriging). Statistical inference is carried out by an empirical Bayes or penalized likelihood approach using mixed model technology. The results confirm that the methodological concept provides useful tools for exploratory analyses of the data at hand and in similar situations.     

Detection and correction of outliers in the bivariate chain-ladder method

The expected profit or loss of a non-life insurance company is determined for the whole of its multiple business lines. This implies the study of the claims reserving problem for a portfolio consisting of several correlated run-off triangles. A popular technique to deal with such a portfolio is the multivariate chain-ladder method of Merz and Wüthrich (2008). However, it is well known that the chain-ladder method is very sensitive to outlying data. For the univariate case, we have already developed a robust version of the chain-ladder method. In this article we propose two techniques to detect and correct outlying values in a bivariate situation. The methodologies are illustrated and compared on real examples from practice.     

Determination of the portfolio selection for a property-liability insurance company

This paper presents an analysis of a portfolio model which can be used to assist a property-liability insurance company in determining the optimal composition of the insurance and investment portfolios. By introducing insurer's threshold risk and relaxing some non-realistic assumptions made in traditional chance constraint insurance and investment portfolio models, we propose a method for an insurer to maximize his return threshold for a given threshold risk level. This proposed model can be used to optimize the composition of underwriting and investment portfolios regarding the insurer's threshold risk level, as well as to generate the efficient frontier by adjusting insurer's threshold risk levels. A numerical example is given based on the industry's aggregated data for a sixteen year period.     

Fair valuation of insurance contracts under Lévy process specifications

The valuation of options embedded in insurance contracts using concepts from financial mathematics (in particular, from option pricing theory), typically referred to as fair valuation, has recently attracted considerable interest in academia as well as among practitioners. The aim of this article is to investigate the valuation of participating and unit-linked life insurance contracts, which are characterized by embedded rate guarantees and bonus distribution rules. In contrast to the existing literature, our approach models the dynamics of the reference portfolio by means of an exponential Lévy process. Our analysis sheds light on the impact of the dynamics of the reference portfolio on the fair contract value for several popular types of insurance policies. Moreover, it helps to assess the potential risk arising from misspecification of the stochastic process driving the reference portfolio.     

Life insurance in a portfolio context

The ownership of life insurance may be modeled as a portfolio problem in which the return on the life insurance contract is negatively correlated with the return on a claim to future wage income. The mean-variance model developed in the paper uses such a framework to express the optimal amount of insurance in terms of two components: the expected value of the wage claim and the risk/return characteristics of the insurance contract. The model thus offers an appealing way to formulate the life insurance problem in a portfolio context. Implications of the model for the functioning of a life insurance market are examined and the existence of accidental death contracts is explained.     

Points systems for car insurance

Some data on car insurance premiums charged to 120 categories of customer, arranged in a 5×4×3×2 table, has been analysed. A method similar to analysis of variance was used to find how closely the data could be reproduced by additive combinations of parameters from the four dimensions, the unusual feature was that these parameters were restricted to be small whole numbers. The reason for this is the attractiveness of such a points system in marketability and in ease of comprehension by the customer.     

Stochastic portfolio specific mortality and the quantification of mortality basis risk

In the last decade a vast literature on stochastic mortality models has been developed. However, these models are often not directly applicable to insurance portfolios because:

(a) For insurers and pension funds it is more relevant to model mortality rates measured in insured amounts instead of measured in the number of policies.

(b) Often there is not enough insurance portfolio specific mortality data available to fit such stochastic mortality models reliably.

Therefore, in this paper a stochastic model is proposed for portfolio specific mortality experience. Combining this stochastic process with a stochastic country population mortality process leads to stochastic portfolio specific mortality rates, measured in insured amounts. The proposed stochastic process is applied to two insurance portfolios, and the impact on the Value at Risk for longevity risk is quantified. Furthermore, the model can be used to quantify the basis risk that remains when hedging portfolio specific mortality risk with instruments of which the payoff depends on population mortality rates.     

Portfolio insurance: Gap risk under conditional multiples

The research on financial portfolio optimization has been originally developed by Markowitz (1952). It has been further extended in many directions, among them the portfolio insurance theory introduced by Leland and Rubinstein (1976) for the “Option Based Portfolio Insurance” (OBPI) and Perold (1986) for the “Constant Proportion Portfolio Insurance” method (CPPI). The recent financial crisis has dramatically emphasized the interest of such portfolio strategies. This paper examines the CPPI method when the multiple is allowed to vary over time. To control the risk of such portfolio management, a quantile approach is introduced together with expected shortfall criteria. In this framework, we provide explicit upper bounds on the multiple as function of past asset returns and volatilities. These values can be statistically estimated from financial data, using for example ARCH type models. We show how the multiple can be chosen in order to satisfy the guarantee condition, at a given level of probability and for various financial market conditions.     

Optimal hedging strategies for multi-period guarantees in the presence of transaction costs: A stochastic programming approach

Multi-period guarantees are often embedded in life insurance contracts. In this paper we consider the problem of hedging these multi-period guarantees in the presence of transaction costs. We derive the hedging strategies for the cheapest hedge portfolio for a multi-period guarantee that with certainty makes the insurance company able to meet the obligations from the insurance policies it has issued. We find that by imposing transaction costs, the insurance company reduces the rebalancing of the hedge portfolio. The cost of establishing the hedge portfolio also increases as the transaction cost increases. For the multi-period guarantee there is a rather large rebalancing of the hedge portfolio as we go from one period to the next. By introducing transaction costs we find the size of this rebalancing to be reduced. Transaction costs may therefore be one possible explanation for why we do not see the insurance companies performing a large rebalancing of their investment portfolio at the end of each year.     

Life insurance policy termination and survivorship

There has been some work, e.g. Carriere (1998), Valdez (2000b), and Valdez (2001), leading to the development of statistical models in understanding the mortality pattern of terminated policies. However, there is a scant literature on the empirical evidence of the true nature of the relationship between survivorship and persistency in life insurance. When a life insurance contract terminates due to voluntary non-payment of premiums, there is a possible hidden cost resulting from mortality antiselection. This refers to the tendency of policyholders who are generally healthy to select against the insurance company by voluntarily terminating their policies. In this article, we explore the empirical results of the survival pattern of terminated policies, using a follow-up study of the mortality of those policies that terminated from a portfolio of life insurance contracts. The data has been obtained from a major insurer which traced the mortality of their policies withdrawn, for purposes of understanding the mortality antiselection, by obtaining their dates of death from the Social Security Administration office. Using a representative sample of this follow-up data, we modeled the time until a policy lapses and its subsequent mortality pattern. We find some evidence of mortality selection and we consequentially examined the financial cost of policy termination.     

Intelligent systems in finance

Business sectors ranging from banking and insurance to retail, are benefiting from a whole new generation of ‘intelligent’ computing techniques. Successful applications include asset forecasting, credit evaluation, fraud detection, portfolio optimization, customer profiling, risk assessment, economic modelling, sales forecasting and retail outlet location. The techniques include expert systems, rule induction, fuzzy logic, neural networks and genetic algorithms, which in many cases are outperforming traditional statistical approaches. Their key features include the ability to recognize and classify patterns, learning from examples, generalization, logical reasoning from premises, adaptability and the ability to handle data which is incomplete, imprecise and noisy. This paper is the first in a series to appear in Applied Mathematical Finance;here we introduce the reader to the basic concepts of intelligent systems, describe their mode of operation and identify applications of the techniques in real world problem domains. Subsequent papers will concentrate on neural networks, genetic algorithms, fuzzy logic and hybrid systems, and will investigate their history and operation more rigorously.     

分形统计测度构建及其在投资组合中的应用研究

准确测量证券的风险和收益无论是对投资管理,还是对金融理论研究,甚至对理论成果向实践应用转化都至关重要。本文在证券价格具有分形特征的现实背景下,基于分形理论构建了分形期望和分形方差两个分形统计测度,以克服非分形统计测度在风险收益方面测不准或不可测的缺陷。在此基础上,应用分形统计测度构建了投资组合模型,给出了分形组合模型的解析解;随后,利用实证分析验证了分形统计测度在投资组合应用中的有效性。本文创新之处在于针对证券价格具有分形特征的现实背景构建了分形期望和分形方差两个分形统计测度;并基于分形统计测度构建了投资组合模型,将证券价格普遍存在的分形特征纳入投资组合的研究框架。     

Multivariate negative binomial models for insurance claim counts

It is no longer uncommon these days to find the need in actuarial practice to model claim counts from multiple types of coverage, such as the ratemaking process for bundled insurance contracts. Since different types of claims are conceivably correlated with each other, the multivariate count regression models that emphasize the dependency among claim types are more helpful for inference and prediction purposes. Motivated by the characteristics of an insurance dataset, we investigate alternative approaches to constructing multivariate count models based on the negative binomial distribution. A classical approach to induce correlation is to employ common shock variables. However, this formulation relies on the NB-I distribution which is restrictive for dispersion modeling. To address these issues, we consider two different methods of modeling multivariate claim counts using copulas. The first one works with the discrete count data directly using a mixture of max-id copulas that allows for flexible pair-wise association as well as tail and global dependence. The second one employs elliptical copulas to join continuitized data while preserving the dependence structure of the original counts. The empirical analysis examines a portfolio of auto insurance policies from a Singapore insurer where claim frequency of three types of claims (third party property damage, own damage, and third party bodily injury) are considered. The results demonstrate the superiority of the copula-based approaches over the common shock model. Finally, we implemented the various models in loss predictive applications.