财产保险公司投资组合问题的多阶段随机规划模型

财产保险公司的投资组合模型均是单期的 ,不能充分满足投资组合管理实践的需要 .为提供多期规划工具 ,建立了一个多阶段的随机规划模型 .它考虑了交易成本 ,分析了不同时期的现金流 ,讨论了资产负债的匹配问题 ,去掉了收益分布的正态假定 ,并增加了一种投资约束 .数值实例的计算结果表明 ,多期模型能更好地帮助财产保险公司选择保险与投资的优化组合 ,其性能要优于单期模型

Multistage Stochastic Programming Model for the Portfolio Problem of a Property-Liability Insurance Company

The current portfolio model for property-liability insurance company is only single period that can not meet the practical demands of portfolio management, and the purpose of this paper is to develop a multiperiod model for its portfolio problem. The model is a multistage stochastic programming which considers transaction costs, cash flow between time periods, and the matching of asset and liability; it does not depend on the assumption for normality of return distribution. Additionally, an investment constraint is added. The numerical example manifests that the multiperiod model can more effectively assist the property-liability insurer to determine the optimal composition of insurance and investment portfolio and outperforms the single period one.