Large portfolio allocation using high-frequency financial data

Asset allocation strategy involves dividing an investment portfolio among different assets according to their risk levels. In recent decades, estimating volatilities of asset returns based on high-frequency data has emerged as a topic of interest in financial econometrics. However, most available methods are not directly applicable when the number of assets involved is large, since small component-wise estimation errors could accumulate to large matrix-wise errors. In this paper, we introduce a method to carry out efficient asset allocation using sparsity-inducing regularization on the realized volatility matrix obtained from intra-day high-frequency data. We illustrate the new method with the high-frequency price data on stocks traded in New York Stock Exchange over a period of six months in 2013. Simulation studies based on popular volatility models are also presented. The proposed methodology is theoretically justified. Numerical results also show that our approach performs well in portfolio allocation by pooling together the strengths of regularization and estimation from a high-frequency finance perspective.