Yugu Xiao, Feng Wu, Zhenzhen Li

【Abstract】

In order to solve the problems of multi-period portfolio, a multi-period investment portfolio optimal process is designed. The process can solve the problems by minimizing the risk with some benefits constraints, using a series of fast algorithm modules. It is achieved by combining the nonparametric sampling method, clustering-based multistage scenario generation method, and multi-period polyhedral risk measures. The process is proposed based on some calculations, which is simple, reasonable and easy to implement. The empirical study on the financial markets data in China demonstrates good practicality of the whole process.

【Keywords】

Multi-stage portfolio, Multifaceted risk measurement, clustering algorithm, Bootstrap; CVaR