状态空间分析在减少工程保险逆向选择中的应用

利用状态空间图首先分析了工程保险市场中逆向选择现象的产生的原因,其次进行了通过保单设计和对投保人进行风险分类以减少逆向选择现象发生的状态空间分析,比较而言,对投保人进行风险分类更加有效,最后得出可以委托监理工程师提供风险分类技术服务的结论.     

基于相关性分析与DEA模型的寿险公司效率分析

选取国内较有影响力的八家寿险公司作为研究对象,提出了基于相关性分析的DEA组合评价方法,该方法综合了相关性分析和DEA两种方法的优点.利用相关性分析的方法设计出评价寿险公司经营效率的投入和产出指标.然后综合运用DEA模型对这八家公司的经营效率进行研究,分析影响效率有效性的因素.     

零膨胀广义泊松回归模型与保险费率厘定

在保险产品的分类费率厘定中,最常使用的模型之一是泊松回归模型.当损失数据存在零膨胀(zero-in flated)特征时,通常会采用零膨胀泊松回归模型.在零膨胀泊松回归模型中,一般假设结构零的比例参数φ为常数,不受费率因子的影响,这有可能背离实际情况.假设参数φ与费率因子之间存在一定关系,并在此基础上建立了零膨胀广义泊松回归模型,即Z IGP(τ)回归模型.通过对一组汽车保险损失数据的拟合表明,Z IGP(τ)回归模型可以有效地改善对实际数据的拟合效果,从而提高费率厘定结果的合理性.     

我国保险企业执行力模糊评价模型与实证分析

执行力不足是现阶段我国保险企业普遍面临的困境.根据保险企业执行力的影响因素,可以建立保险企业执行力评价指标体系,并采用AHP与模糊评价相结合的方法对我国保险企业执行力进行模糊评价.各保险企业可以从评价结果中找出自身执行力的薄弱环节加以改进,以提升企业执行力。     

Minimum standards for investment performance: A new perspective on non-life insurer solvency

The aim of this paper is to develop an alternative approach for assessing an insurer’s solvency as a proposal for a standard model for Solvency II. Instead of deriving minimum capital requirements–as is done in solvency regulation–our model provides company-specific minimum standards for risk and return of investment performance, given the distribution structure of liabilities and a predefined safety level. The idea behind this approach is that in a situation of weak solvency, an insurer’s asset allocation can be adjusted much more easily in the short term than can, for example, claims cost distributions, operating expenses, or equity capital. Hence, instead of using separate models for capital regulation and solvency regulation–as is typically done in most insurance markets–our single model will reduce the complexity and costs for insurers as well as for regulators. In this paper, we first develop the model framework and second test its applicability using data from a German non-life insurer.     

Securitization of motor insurance loss rate risks

In an attempt to transfer the loss rate risks in motor insurance to the capital market, we use the tranche technique to hedge the motor insurance risks. This paper illustrates AXA and their securitization of French motor insurance in 2005 as an example. Though this application is new, this transaction is based on a concept similar to CDOs. Tranches of bonds are constructed on the basis of the expected loss ratio from motor insurance policy holders’ groups. As a consequence we develop motor loss rate bonds using the structure of synthetic CDOs. The coupon payments of each tranche depend on the level of the loss rates of the underlying motor insurance pool. We show the integral formulas for the loss tranche contract where the loss distribution is modelled with discounted compound Poisson process. Esscher transform is chosen for a risk adjusted measure change and Fourier inversion method is used to calculate the price of the motor claim rate securities. The pricing methods of the tranches are illustrated, and possible suggestions to improve the pricing method and the design of these new securities follow.     

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。