An Evolutionary Computation Approach to Scenario-Based Risk-Return Portfolio Optimization for General Risk Measures
An Evolutionary Computation Approach to Scenario-Based Risk-Return Portfolio Optimization for General Risk Measures
复制标题
一般风险度量的基于场景的风险回报投资组合优化的进化计算方法
DOI：
10.1007/978-3-540-71805-5_22
复制
发表时间：
2009
期刊：
影响因子：
--
通讯作者：
Ronald Hochreiter
中科院分区：
文献类型：
--
作者：
Ronald Hochreiter
由于金融工程问题的复杂性和非凸性日益增加，生物启发的启发式算法在金融决策优化领域具有重要意义。本文分析了基于随机场景的风险收益组合优化问题，并用进化计算方法求解了该问题。应用这种方法的好处是为任意一组基于损失分配的风险度量创建了一个通用框架，而不管它们的底层结构如何。本文总结了三种最常用的风险度量方法的数值结果。
Due to increasing complexity and non-convexity of financial engineering problems, biologically inspired heuristic algorithms gained significant importance especially in the area of financial decision optimization. In this paper, the stochastic scenario-based risk-return portfolio optimization problem is analyzed and solved with an evolutionary computation approach. The advantage of applying this approach is the creation of a common framework for an arbitrary set of loss distribution-based risk measures, regardless of their underlying structure. Numerical results for three of the most commonly used risk measures conclude the paper.