Abstract: It is of great theoretical value and practical significance to construct appropriate investment strategies to better deal with the impact of extreme risk events by taking the possibility of the occurrence of future extreme losses into account in the investment decision-making process. This paper builds and seeks solution to the asset portfolio selection model on the basis of extreme loss constraints. It conducts a numerical and empirical analysis by using the data of monthly return rateof the ten industry indexes of HS300 from January 2005 to August 2020, which extends the research of asset portfolio selection based on the underneath risk control. The result shows that the efficient portfolio based on the extreme loss constraints satisfies the K+2 fund separation theorem.The extreme loss constraints can increase the skewness of efficient portfolio returns at the expense of mean variance efficiency.The extreme loss constraints can also improve the out-of-sample average return and the Sharpe ratio of the minimum variance portfolio (MVP); the MVP with extreme loss constraints can deliver a higher net Sharpe ratio than the equal weight portfolio and the traditional MVP even if the transaction cost be taken into consideration. On account of this, the investors can consider to take the extreme loss constraints into the investment decision-making process, so as to obtain better investment performance with risk control.

Key words: extreme loss constraints, asset portfolio selection, portfolio performance, stock market