Research On High Frequency Trading Strategy Portfolio Model And Its Application

The high-speed development of high-frequency trading overseas and deepening reform of financial market domestic bring the new chance for booming of the Chinese high-frequency trading, which poses the request to research on the management of high-frequency trading strategy further. The mean-variance model proposed originally by Markowitz, has played an important role in the development of modern portfolio selection theory. However, under the high-frequency environment, the traditional Markowitz model was no longer suitable for high-frequency trading data. In this paper, we did a lot of work on theoretical research, model design, algorithm design and empirical study. The main studies are as follows. 1. Considering the high demand of high-frequency environment on the accuracy of data, we added “risk aversion” based on the traditional Markowitz model. On the covariance estimation, based on the Hayashi-Yoshida estimator, we redesigned the covariance estimator, and then used it as one of the input parameters. Meanwhile, this paper introduced the idea of dynamic management, and we used the standard artificial bee colony algorithm to solve the modified portfolio model to verify the validity of the model. 2. Furthermore, in this paper, we tried to choose strategies which perform well from a lot of high frequency trading strategies, and used these strategies to build portfolio for the purpose of realizing the revenue maximization. Therefore, we brought in the multi-factor model, choose the appropriate evaluation indicators for high frequency trading strategies and then applied the idea of multifactor model to choose strategies. At last, we applied the modified artificial bee colony algorithm to solve the high-frequency trading strategy portfolio. 3. We build the quantitative environment for empirical analysis—multi-factor model, high-frequency trading strategy portfolio model based on modified Artificial Bee Colony(ABC) algorithm. 4. Finally, this paper illustrated that dynamic adjusting can improve portfolio returns. Further, it proved the important of choosing strategies dynamically. In conclusion, this paper applied theory knowledge to practice and it realize the combination of theory and practice. It has important theoretical significance and application prospects.