A0705
Title: Multiperiod dynamic portfolio choice: When high dimensionality meets return predictability
Authors: Wenfeng He - Renmin University of China (China) [presenting]
Mei Xiaoling - Xiamen University (China)
Wei Zhong - Xiamen University (China)
Huanjun Zhu - Xiamen University (China)
Abstract: A novel two-step methodology is developed to solve the multiperiod dynamic portfolio choice problem with high dimensional assets in the presence of return predictability conditional on a large number of state predictors. Specifically, in the first step, the new risk-premium projected-PCA (RP-PPCA) method is proposed to reduce the dimension of tradable assets. This method achieves dimension reduction (DR) by estimating latent factors with explanatory power in time series variation and expected return in high-dimension-low-sample-size data. In the second step, dynamic programming is used to solve the multiperiod portfolio choice problem. In each recursive step, an adjusted semiparametric model averaging (AMA) method is adopted to avoid the curse of dimensionality associated with a large set of state variables while remaining computationally efficient. Thus, the two-step approach is named DRAMA, which stands for a combination of a new dimension reduction method and an adjusted semiparametric model averaging method. Analytically, it is shown that the portfolios constructed by the DRAMA are approximately optimal under mild assumptions. Moreover, the numerical results based on empirical data from US stock markets show that the proposed portfolios have excellent out-of-sample performances.
