MULTI-PERIOD DYNAMIC PORTFOLIO OPTIMIZATION THROUGH LEAST SQUARES LEARNING
MULTI-PERIOD DYNAMIC PORTFOLIO OPTIMIZATION THROUGH LEAST SQUARES LEARNING
复制标题
通过最小二乘学习进行多周期动态投资组合优化
DOI：
10.1142/9789814667364_0003
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Geoffrey Lee
中科院分区：
文献类型：
--
作者：
Chenming Bao;Z. Zhu;N. Langrené;Geoffrey Lee
本文描述了一种求解动态投资组合选择问题的算法。将投资组合选择问题建模为多重转换问题，并实现了一种基于仿真的数值方法求解动态投资组合优化问题。基于最小二乘蒙特卡罗方法的递归数值方法被用来计算离散决策日期序列的投资者的条件价值函数。该方法是数据驱动的，不限于特定的资产模型。重要的是，在动态优化过程中考虑了与投资组合再平衡相关的中间交易成本。在目前的实施中，也考虑到了投资者的风险偏好和风险管理约束。本文介绍了一个投资于五个股票市场的全球股票组合的案例研究，其中还包括外汇风险。案例研究提供了一个数值例子，使用的方法，为8维。
This paper describes an algorithm to solve a dynamic portfolio selection problem. The portfolio selection problem is modelled as multiple switching problem, and a simulation-based numerical method is implemented for solving the dynamic portfolio optimization problem. A recursive numerical approach based on the Least Squares Monte Carlo method is used to calculate the conditional value functions of investors for a sequence of discrete decision dates. The methodology is data driven, is not restricted to specific asset models. Importantly, intermediate transaction costs associated with portfolio rebalancing is considered in the dynamic optimisation process. Investors' risk preferences and risk management constraints are also taken into account in the current implementation. A case study is presented for a global equity portfolio invested in five equity markets, and foreign exchange risks are also included. The case study provides a numerical example of using the methodology for 8-dimensions.