The Optimal Dynamic Portfolio Model Based on High-Frequency Data
Yu Xing · International Journal of Applied Mathematics & Statistics/International journal of applied mathematics and statistics · 2013
The objective of this research is to develop a dynamic portfolio selection model based on high-frequency data, in which the objective is to minimax the portfolio risk described as realized covariance matrix under the constraints of return, budget, transaction costs etc.This paper forecasts the realized volatility and the correlation between two risk assets using ARMA process. And based on the forecasting result to construct the optimal portfolio model, the transformed model can be solved by an optimization technique. An empirical study is used to illustrate the proposed approach and gives an investor some investing suggestions.