政策约束下基于不同风险测度的最优保险投资策略

This proposal focuses on employing several risk measures, constructing new insurance investment models and finding the optimal strategies under government policy constraints based on China’s insurance investment practice and the limit of existing references. The main modeling idea is developed from the following three aspects, (1) constructing new risk adjusted return of capital (RAROC) based on the character of risk and risk attitude, (2) constructing insurance investment model with two risk measures and (3) introducing systemic risk measures based on risk linkage and contagion. In view of the above three points, different risk surplus models ，different distributions and dependence of return, a variety of models are built and solved. The effect of policy change, risk attitude, risk measures and other factors on optimal insurance portfolio will be explored and these models’ applicability will be discussed. This research needs to synthesize existing research methods and the new knowledge in the fields of stochastic analysis, optimization theory, asset fluctuation theory and risk estimation methods etc. and make continuous innovations to obtain meaningful conclusions by comparison from the point of views of theoretical study, simulation and real examples. The proposed research results will not only develop and improve insurance portfolio theory, but also provide practically alternative scientific decision-making thinking clues and methods for the managers in insurance company and other financial institutions.

本项目基于我国保险投资现状及已有研究的局限性, 采用多种风险度量方法, 考虑到国家政策约束, 建立新的保险投资模型, 探讨最优保险投资策略。主要建模思想有（1）考虑风险特征和投资者态度，构造新的风险调整报酬率；（2）兼顾不同风险特征，构建多重风险优化模型；（3）考虑风险的联动性和传染性，引入系统性风险指标。研究内容：分别基于上述三种建模思想，在不同承保风险模型、不同收益率分布和相依关系下建立多种优化模型并求解，探讨政策变化、投资态度、风险测度方法等对最优投资策略的影响，分析模型适用性。研究工具：以此方面已有研究为基础，借鉴随机分析、最优化、风险估计等领域新理论和方法，不断融合创新，从理论、模拟、实证角度进行对比研究，以获得有意义结果。研究结果理论上可发展完善已有保险投资决策理论，实践上可为保险公司乃至其它金融机构的投资管理者提供可供选择的科学决策思路和方法。

风险度量是投资策略研究中的核心问题，随着经济、金融及社会大环境的快速发展，风险呈现更加复杂多变的状态，这对投资组合模型的理论建模提出了新的要求。本项目分别从风险态度、风险的多样性、风险的系统性三个角度出发，构建了多种投资组合优化模型，研究最优保险（金融）投资策略问题。项目研究贡献主要体现在：（1）从风险态度出发，利用谱风险测度、CARA效用函数、基于CVaR的尾部风险厌恶指标等刻画投资主体的态度，研究最优保险投资及金融投资策略问题。（2）从风险多样性出发，在均值-方差-CVaR优化准则下，探讨最优保险投资及投资-再保险策略问题，进一步在高维数据、尾部风险厌恶态度、系统风险、模糊风险等背景下对金融投资模型和优化路径进行创新探索。（3）从系统风险角度，利用网络理论，构建新的金融网络结构及中心性指标，结合多种降噪工具，研究最优投资策策问题。基于以上三个角度的研究，理论模型是全新的，模拟实证分析在多种评价准则下都有良好绩效表现。同时，项目也对基准风险、信度风险测度（CrVaR）、条件在险价值（CoVaR）、CTE、扭曲风险测度等的估计及在金融（保险）投资策略、风险溢出效应、精算定价应用进行深入探索，对中国保险业减贫效应、巨灾保险及农产品保险等问题进行了研究。项目研究结果不仅能够丰富投资组合、风险测度、网络理论等的理论体系，而且能为保险、金融行业的资产管理及风险分析技术提供更加多样、可行的数量分析工具和理论依据。