农业保险“茶艺”

得到精确的茶叶单产分布对准确厘定费率很重要，而单产分布则与茶叶种类、植茶区域、茶园年龄、茶园小气候等均有关系。     

运用时空模型厘定湖北省县级水稻产量保险的纯费率

本文建立了23个综合考虑水稻单产的时间效应、空间效应和时空交互效应的时空模型。利用湖北省1991-2007年县级水稻单产数据,借助WinBUGS进行Gibbs抽样估计,根据DIC准则进行选择最优模型,预测1992-2009年各县市的水稻单产。基于2008年预测单产及其后验分布,厘定县级水稻产量保险的纯费率。研究的主要结论:1992-2007年的预测值与实际值比较接近,且预测的水稻单产的标准差都比较小,表明所选择模型具有良好的短期预测能力;相邻地域的纯费率比较接近;所厘定的纯费率与蒙特卡洛(Monte Carlo,MC)误差正相关,Pearson相关系数为0.6793,MC误差包含来自时间、空间以及两者相互作用带来的不确定性。     

一些最大信息先验分布与广义最大熵先验分布

本文讨论了三种确定无信息先验分布的方法。利用熵不等式,简化了Zellnor(1984)关于最大数据信息先验分布的一条定理的证明,并得到了其唯一性。在此基础上提出了广义最大熵先验分布,它在某些方面改进了经典的最大熵原理,对于最大相关先验分布(即Lindley准则),我们得到了一个中间解,由此导出了许多常见分布族的最大相关先验分布。     

市场约束条件下的非寿险费率厘定

在非寿险费率厘定中,经常遇到的一个实际问题是某些风险类别的费率不能过高或不能过低。在这种约束条件下,传统的广义线性模型将不能直接用于费率厘定。本文给出了一种在一般线性约束条件下,如何应用迭代算法对常用的广义线性模型进行调整,从而得到满足特定约束条件的费率厘定结果。本文的实证研究结果表明,该方法具有灵活性和现实可行性,能够解决非寿险费率厘定中常见的市场约束问题。     

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。     

奖惩系统的数学建模与稳态分析

奖惩系统(Bonus- Malus Systems)是世界各国机动车辆险中广泛采用的一种经验费率厘定机制.本文在最一般的框架下,以有限齐次马尔科夫链对奖惩系统建模,证明了任一奖惩系统皆存在唯一的平稳分布,此外,给出了求解这一平稳分布的一般算法,并揭示了此平稳分布的结构.特别地,针对两种具有特定转移法则的奖惩系统,还给出了它们平稳分布的简明显式表示.     

车险费率厘定精算技术的研究与应用评述

精算技术为中国车险市场费率改革提供必要支持,可以确保费率厘定的科学性与合理性。首先,本文系统梳理了车险分类风险费率厘定精算统计模型的发展历程,并回顾参数估计方法。其次,论述了车险个体风险费率厘定的精算模型与方法,并重点评述了信度理论与奖惩系统的研究。进而,归纳出车险费率厘定精算统计模型的研究热点与发展方向。最后,指明现有研究对中国车险费率厘定精算方法的启示,并提出相关建议。     

汽车交通事故损失分布的尾部估计

由于巨额损失对保险公司的影响非常大,费率厘定过程中对极值分布的研究非常重视。本文从极值理论的角度出发,以上海市虹口区2003年的汽车交通事故损失数据为样本,探讨了损失分布的尾部估计方法,并利用该地区2006年的汽车交通事故损失对结论进行了验证。研究结果发现,广义帕雷托分布确实对损失额的尾部提供了较好的拟合,但这依赖于门槛值的恰当选择。传统的门槛值选择方法主观性较强,而通过重复多次的交叉验证技巧,我们可以估计广义帕雷托分布的最合适门槛值位置。     

基于极大熵准则的先验分布确定方法

借助于熵的概念，讨论了用极大熵的思想来确定先验分布的几种情形，给出了在各种情形下先验分布的形式和结论，从而提供了确定先验分布的一种有效的方法。     

不完全信息下损失风险的研究

在获得损失分布不完全信息情况下,提出用方差和熵共同度量损失风险的方法.在不完全信息条件下,通过最大熵原理在最不确定的情况下得到最大熵损失分布,并获得了损失分布的熵函数值.用熵值度量损失分布对于均匀分布的离散程度,从而度量概率波动带来的风险;用方差度量损失对于均值的离散程度,从而度量状态波动带来的风险.由于熵是与损失变量更高阶矩信息相联系的,所以新方法是从更全面的角度对损失风险的预测.通过算例,进一步看出在获得高阶矩信息下,熵参与风险度量的必要性.     

最大熵——均值方差保费原则

本为利用熵在金融市场的两个功能;度量风险资产的投资风险和推测资产的概率分布,抓住了不确定性的本质,用熵值来度量由概率分布向信息转化的不确定性,建立了新的保费原则;最大熵—均值方差保费原则,使保费的制定更趋于合理.     

指数保费原理下的经验厘定

在经典的信度理论中,信度保费是在净保费原理下得到的. 但是, 保险商业中, 保险公司要求制定的保费必须适用于某合适的保费原理以适应具体的保险商业的需要. 本文建立了指数保费原理下的完全经验厘定模型, 得到了风险保费的信度估计和经验Bayes 信度估计, 并讨论了结构参数的估计及其性质. 最后证明了多合同模型的经验Bayes 信度估计的渐近最优性     

The Log–Lindley distribution as an alternative to the beta regression model with applications in insurance

In this paper a new probability density function with bounded domain is presented. The new distribution arises from the generalized Lindley distribution proposed by Zakerzadeh and Dolati (2010). This new distribution that depends on two parameters can be considered as an alternative to the classical beta distribution. It presents the advantage of not including any special function in its formulation. After studying its most important properties, some useful results regarding insurance and inventory management applications are obtained. In particular, in insurance, we suggest a special class of distorted premium principles based on this distribution and we compare it with the well-known power dual premium principle. Since the mean of the new distribution can be normalized to give a simple parameter, this new model is appropriate to be used as a regression model when the response is bounded, being therefore an alternative to the beta regression model recently proposed in the statistical literature.     

构造和评价奖惩系统的新方法

当知道一个奖惩系统的等级和保费水平时,利用信息熵提出并得到奖惩系统的理想概率分布．在保单中考虑人身伤亡索赔次数的二元模型,将奖惩系统的平稳概率分布与理想概率分布的差异,通过卡方值与多余度分别表现出来,将其作为评价奖惩系统的一个新工具．     

GAMLSS模型在车险定价中的应用

在系统梳理国内外非寿险产品费率厘定方法的基础上,详细介绍了GAMLSS模型,证明了在位置参数和尺度参数的预测中均引入随机效应的GAMLSS模型可更有效地解释纵向数据中个体间的异质性.最后将GAMLSS模型应用于一组纵向车辆保险数据,计算了先验保费、后验保费、后验风险保费和奖惩因子.实证结果表明,GAMLSS模型不仅可为非寿险产品的定价提供依据,而且使风险分类更加稳定、合理.     

Generalizations of common ILF models

In Germany (and various other countries) a specific ILF model, which was proposed by Riebesell in 1936, is frequently used for exposure rating in liability insurance and reinsurance. This model is based on the assumption that the doubled limits surcharge (DLS) – i.e. the percentage surcharge on the premium for doubling policy limits – does not depend on the policy limit. The drawback of this method is its lack of flexibility with respect to the corresponding severity distribution. In this paper we analyze the following generalizations of this concept which offer more flexibility regarding the severity: stepwise reduction of the DLS, continuously decreasing DLSs and construction of an ILF formula based on the DLS at a certain policy limit and the tail of a severity distribution.     

平衡指数损失函数下的信度保费

在经典的信度保费模型中,得到的信度保费估计均是考虑的是纯保费,然而在保险实务中,保险公司收取的保费不可能是纯保费,必须具有正的安全负荷.在平衡指数损失函数下,研究了多合同的信度保费模型.利用正交投影方法,得到了未来保费的信度估计.最后对估计进行了数值模拟.     

Simulation of a health insurance market with adverse selection

A health insurance market is examined in which individuals with a history of high utilization of health care services tend to select fee-for-service (FFS) insurance when offered a choice between FFS and health maintenance organizations (HMOs). In addition, HMOs are assumed to practice community rating of employee groups. Based on these observations and health plan enrollment and premium data from Minneapolis-St. Paul, a deterministic simulation model is constructed to predict equilibrium market shares and premiums for HMO and FFS insurers within a firm. Despite the fact that favorable selection enhances their ability to compete with FFS insurers, the model predicts that HMOs maximize profits at less than 100% market share, and at a lower share than they could conceivably capture. That is, HMOs would not find it to their advantage to drive FFS insurers from the market even if they could. In all cases, however, the profit-maximizing HMO premium is greater than the experience-rated premium and, thus, the average health insurance premium per employee in firms offering both HMOs and FFS insurance is predicted to be greater than in firms offering one experience-rated plan. The model may be used to simulate the effects of varying the employer's method of contributing to health insurance premiums. Several contribution methods are compared. Employers who offer FFS and HMO insurance and pay the full cost of the lowest-cost plan are predicted to have lower average total premiums (employer plus employee contributions) than employers who pay any level percent of the cost of each plan.     

The credibility premiums for exponential principle

In the classical credibility theory, the credibility premium is derived on the basis of pure premium. However, the insurance practice demands that the premium must be charged under some adaptable premium principle and serves the purpose for insurance business. In this paper, the balanced credibility models have been built under exponential principle, and the credibility estimator of individual exponential premium is derived. This result is also extended to the versions of multitude contracts, and the estimation of the structure parameters is investigated. Finally, the simulations have been introduced to show the consistency of the credibility estimator and its differences from the classical one.