考虑空间效应的贝叶斯分层模型与索赔频率预测

在非寿险索赔频率预测中,使用最为广泛的是广义线性模型.但是,如果观察数据呈现出明显的零膨胀特征,或者包含空间协变量,或者某些协变量之间具有分层结构,则广义线性模型的拟合优度往往欠佳.在零膨胀分布假设下,建立了考虑空间效应的贝叶斯分层模型,并将其应用于索赔频率预测.在模型中,用惩罚样条函数描述连续型协变量的非线性效应,用高斯马尔科夫随机场描述相邻地区在索赔频率上的空间相依性,用随机截距项描述不同地区在索赔频率上的分层关系和差异性.实证研究结果表明,考虑空间效应的贝叶斯分层模型的拟合优度明显优于传统的广义线性模型.     

混合效应模型及其在非寿险费率厘定中的应用

在非寿险分类费率厘定中,广义线性模型的应用十分普遍,但当某些费率因子的水平数很多时(本文称之为多水平因子),广义线性模型的估计结果将不可靠。解决此类问题的一种方法是把多水平费率因子作为随机效应处理。将多水平费率因子作为随机效应处理可以采取下述三种方法:(1)分别用广义线性模型和信度模型估计普通费率因子和多水平因子,通过广义线性模型与Buhlmann-Straub信度模型的迭代应用预测索赔频率和索赔强度;(2)应用广义线性混合模型分别预测索赔频率和索赔强度;(3)直接对经验纯保费数据建立Tweedie混合效应模型。本文把上述模型应用于中国车损险实际数据的研究结果表明,这三种方法比较接近,但从总体上看,广义线性混合模型的估计结果更加可取。     

GAMLSS模型及其在车损险费率厘定中的应用

在车损险费率厘定中,通常假设索赔频率、索赔强度或纯保费服从指数分布族,并对其均值建立广义线性模型,而假设其他参数对所有风险类别都是固定的常数。这种假设在某些情况下并非成立。GAMLSS模型可以在各种分布假设下同时对一个分布的位置参数、尺度参数和形状参数建立参数或非参数的回归模型,具有很大的灵活性。本文在零调整逆高斯分布假设下把GAMLSS模型应用于我国实际的车损险数据,建立了车损险的费率厘定模型,结果表明,这种模型对车损险实际数据的拟合要优于常用的Tweedie分布假设下的广义线性模型。此外,这种模型厘定的风险保费更加公平合理。     

异方差的线性混合模型的参数估计

本文研究既含有固定效应又含有随机效应的线性混合模型,在随机效应的方差不同即异方差情况下,即考虑方差受外界因素的影响,如温度、湿度等,我们称之为协变量,在有协变量情况下对方差建立对数线性模型,运用最大似然估计讨论了固定效应的估计和随机效应的预测,并且用约束最大似然(REML)方法研究对数线性模型中参数和随机误差中参数(离差参数)的估计,并讨论估计量的性质及离差参数估计量的渐近正态性。     

联合广义线性模型中的变量选择

在联合广义线性模型中,均值和散度参数都被赋予了广义线性模型的结构,本文主要考虑该模型的变量选择问题. 文章利用扩展拟似然函数,提出了一个适用于联合广义线性模型的新的变量选择准则(EAIC),该准则是Akaike信息准则的推广.论文通过模拟研究和一个实例分析验证了该准则的效果.     

考虑个体保单风险特征的最优奖惩系统

奖惩系统在汽车保险中的应用非常普遍。论文首先介绍和讨论了泊松-伽马假设下的最优奖惩系统及其性质;其次在假设个体保单的索赔频率服从二项分布,而二项分布的一个参数服从贝塔分布的条件下,建立了一种考虑个体保单风险特征信息的最优奖惩系统,其中风险特征信息可以通过广义线性模型的形式引入奖惩系统;然后在假设个体保单的索赔频率服从负二项分布,而负二项分布的一个参数服从贝塔分布的条件下,建立了另一个最优奖惩系统;最后讨论了这两个奖惩系统的性质和应用。     

零调整分位数回归模型在车辆保险索赔额中的研究与应用

车辆保险产品的定价一般会考虑保单持有人的索赔概率和期望索赔额等两个因素,零调整逆高斯回归模型作为解决这类问题的一个有力工具,由于变量分布的限定,从而具有一定的局限性.针对该问题,本文基于零调整逆高斯回归模型和分位数回归模型的思想,提出零调整分位数回归模型,并结合实际数据进行了拟合分析.与零调整逆高斯回归模型拟合的结果比较表明,零调整分位数回归模型可以作为研究车辆保险中索赔额的一个有力工具.     

分层广义线性模型在准备金评估中的建模研究

考虑到赔付流量三角形数据同一事故年反复观测的纵向特征以及数据结构的层次性,建立了分层广义线性模型.与通常的随机模型相比,分层广义线性模型不但可以选择条件反应变量的分布而且风险参数分布范围也更加广泛.利用h-似然函数估计分层广义线性模型的模型参数,降低了计算量.为使模型具有可比性,评估模型的预测精度,推导了模型预测误差的估计式.为充分利用已知赔付信息,将赔付额和赔付次数两种赔付信息纳入未决赔款准备金评估模型,建立了两阶段分层广义线性模型.在线性预测量中考虑了各种固定效应和随机效应以及模型结构的散布参数,改进了线性预估量结构.研究表明:分层广义线性模型对于数据的各种分布及形式都具有很好的适应性,更加符合保险实务现实的赔付规律.     

指数族广义非线性随机系数模型的变离差检验

指数族广义非线性随机系数模型是Smith &; Heitjan［１０］和 Wei et al［１１］所研究模型的推广。该文分别在模型离差 (dispersion) 的权不变和变异时，讨论了指数族 广义非线性随机系数模型的变离差的检验问题，得到了score检验统计量。并利用欧洲野兔数据，分别对正态分布模型、Γ 分布模型和 逆高斯分布模型说明检验方法的有效性。     

基于分位广义双曲分布的联合建模

众所周知,广义双曲(generalized hyperbolic,GH)分布因具有比高斯分布更好的拟合而在金融时间序列建模方面有着广泛的应用,例如用GH分布拟合观察到的对数收益数据,因为高斯分布不能捕捉到外汇汇率对数收益标准化后的极端值的半重尾性质.然而,在实践中我们很少使用广义双曲分布,因为很难同时有效地得到5个参数的估计.为了克服这个困难,我们对响应变量服从广义双曲分布的数据提出了一种新的联合建模的方法,其中参数可以通过协变量的简单线性和对数线性形式进行建模.此外,我们分别用使用EM算法和鞍点逼近方法来对参数和分位数进行估计.并证明了分位数估计量的相合性和渐近正态性.     

Severity modeling of extreme insurance claims for tariffication

Generalized linear models are common instruments for the pricing of non-life insurance contracts. They are used to estimate the expected frequency and severity of insurance claims. However, these models do not work adequately for extreme claim sizes. To accommodate for these extreme claim sizes, we develop the threshold severity model, that splits the claim size distribution in areas below and above a given threshold. More specifically, the extreme insurance claims above the threshold are modeled in the sense of the peaks-over-threshold methodology from extreme value theory using the generalized Pareto distribution for the excess distribution, and the claims below the threshold are captured by a generalized linear model based on the truncated gamma distribution. Subsequently, we develop the corresponding concrete log-likelihood functions above and below the threshold. Moreover, in the presence of simulated extreme claim sizes following a log-normal as well as Burr Type XII distribution, we demonstrate the superiority of the threshold severity model compared to the commonly used generalized linear model based on the gamma distribution.     

基于藤Copula的GAMLSS模型与非寿险准备金评估

在假设各个业务线的增量已决赔款服从伽玛分布、逆高斯分布和对数正态分布的基础上,建立了各个业务线增量已决赔款的GAMLSS模型,并将此模型应用于一组具有明显异方差的车险数据,拟合效果优于均值回归模型.另外,在多个业务线的准备金估计中,不同业务线之间的相依性通过藤Copula函数来描述.用D藤Copula描述相依关系的GAMLSS模型对准备金的评估结果既优于独立假设下的GAMLSS模型和链梯法对准备金的评估结果,同时还刻画了不同业务线之间的尾部相依性.     

Generalized linear models for dependent frequency and severity of insurance claims

Traditionally, claim counts and amounts are assumed to be independent in non-life insurance. This paper explores how this often unwarranted assumption can be relaxed in a simple way while incorporating rating factors into the model. The approach consists of fitting generalized linear models to the marginal frequency and the conditional severity components of the total claim cost; dependence between them is induced by treating the number of claims as a covariate in the model for the average claim size. In addition to being easy to implement, this modeling strategy has the advantage that when Poisson counts are assumed together with a log-link for the conditional severity model, the resulting pure premium is the product of a marginal mean frequency, a modified marginal mean severity, and an easily interpreted correction term that reflects the dependence. The approach is illustrated through simulations and applied to a Canadian automobile insurance dataset.     

A Bayesian nonparametric model and its application in insurance loss prediction

Predicting insurance losses is an eternal focus of actuarial science in the insurance sector. Due to the existence of complicated features such as skewness, heavy tail, and multi-modality, traditional parametric models are often inadequate to describe the distribution of losses, calling for a mature application of Bayesian methods. In this study we explore a Gaussian mixture model based on Dirichlet process priors. Using three automobile insurance datasets, we employ the probit stick-breaking method to incorporate the effect of covariates into the weight of the mixture component, improve its hierarchical structure, and propose a Bayesian nonparametric model that can identify the unique regression pattern of different samples. Moreover, an advanced updating algorithm of slice sampling is integrated to apply an improved approximation to the infinite mixture model. We compare our framework with four common regression techniques: three generalized linear models and a dependent Dirichlet process ANOVA model. The empirical results show that the proposed framework flexibly characterizes the actual loss distribution in the insurance datasets and demonstrates superior performance in the accuracy of data fitting and extrapolating predictions, thus greatly extending the application of Bayesian methods in the insurance sector.     

商业医疗保险损失分析:基于广义线性模型的实证研究

本文使用广义线性模型对商业医疗保险损失进行建模,并用某商业保险公司的医疗保险赔付数据进行了实证检验,结果表明,在影响医疗保险损失的诸多因素中,住院天数、医院级别、地区、保障档次等都是显著的因素,而性别和小于60岁以下年龄段内年龄则并不是显著因素,这些结论给医疗保险的经营和风险控制带来实际的意义.     

Bayesian modelling of financial guarantee insurance

In this paper we model the claim process of financial guarantee insurance, and predict the pure premium and the required amount of risk capital. The data used are from the financial guarantee system of the Finnish statutory pension scheme. The losses in financial guarantee insurance may be devastating during an economic depression (i.e., deep recession). This indicates that the economic business cycle, and in particular depressions, must be taken into account in modelling the claim amounts in financial guarantee insurance. A Markov regime-switching model is used to predict the frequency and severity of future depression periods. The claim amounts are predicted using a transfer function model where the predicted growth rate of the real GNP is an explanatory variable. The pure premium and initial risk reserve are evaluated on the basis of the predictive distribution of claim amounts. Bayesian methods are applied throughout the modelling process. For example, estimation is based on posterior simulation with the Gibbs sampler, and model adequacy is assessed by posterior predictive checking. Simulation results show that the required amount of risk capital is high, even though depressions are an infrequent phenomenon.     

Numerical algorithms for Panjer recursion by applying Bernstein approximation

In actuarial science, Panjer recursion (1981) is used in insurance to compute the loss distribution of the compound risk models. When the severity distribution is continuous with density function, numerical calculation for the compound distribution by applying Panjer recursion will involve an approximation of the integration. In order to simplify the numerical algorithms, we apply Bernstein approximation for the continuous severity distribution function and obtain approximated recursive equations, which are used for computing the approximated values of the compound distribution. The theoretical error bound for the approximation is also obtained. Numerical results show that our algorithm provides reliable results.     

广义线性模型在汽车延保风险保费测算中的应用

基于保险公司2010年1月—2019年3月的实际保单数据样本,分别运用广义线性模型中的泊松模型和伽玛模型测算出险频率和案均赔款,构建风险保费测算模型,对影响风险保费的因素进行定量研究及分析.结果表明:该方法能够构建多个变量与风险保费的数值关系,减少了信息的损失,得到的费率表可作为实际应用的参考.最后,通过该方法测算结果与市场定价的实例比较对方法的合理性与优越性进行了说明.     

基于伽玛与对数正态分布假设下的广义线性模型的比较和应用

本文在假设每次损失金额的变异系数相同,且它们服从伽玛分布或对数正态分布的条件下,讨论了加法模型和乘法模型的参数估计和拟合优度检验,并应用一组实际损失数据对上述模型进行了实证比较。结果表明,对于一组特定的损失数据,对数正态分布假设下的广义线性模型可能优于伽玛分布假设下的广义线性模型。     

Dynamical insurance models with investment: Constrained singular problems for integrodifferential equations

Previous and new results are used to compare two mathematical insurance models with identical insurance company strategies in a financial market, namely, when the entire current surplus or its constant fraction is invested in risky assets (stocks), while the rest of the surplus is invested in a risk-free asset (bank account). Model I is the classical Cramér–Lundberg risk model with an exponential claim size distribution. Model II is a modification of the classical risk model (risk process with stochastic premiums) with exponential distributions of claim and premium sizes. For the survival probability of an insurance company over infinite time (as a function of its initial surplus), there arise singular problems for second-order linear integrodifferential equations (IDEs) defined on a semiinfinite interval and having nonintegrable singularities at zero: model I leads to a singular constrained initial value problem for an IDE with a Volterra integral operator, while II model leads to a more complicated nonlocal constrained problem for an IDE with a non-Volterra integral operator. A brief overview of previous results for these two problems depending on several positive parameters is given, and new results are presented. Additional results are concerned with the formulation, analysis, and numerical study of “degenerate” problems for both models, i.e., problems in which some of the IDE parameters vanish; moreover, passages to the limit with respect to the parameters through which we proceed from the original problems to the degenerate ones are singular for small and/or large argument values. Such problems are of mathematical and practical interest in themselves. Along with insurance models without investment, they describe the case of surplus completely invested in risk-free assets, as well as some noninsurance models of surplus dynamics, for example, charity-type models.