A0276
Title: Diversifying risk parity portfolios with high-frequency principal components
Authors: Massimiliano Caporin - University of Padova (Italy)
Laura Garcia-Jorcano - Universidad de Castilla-La Mancha (Spain) [presenting]
Juan-Angel Jimenez-Martin - Complutense of Madrid (Spain)
Abstract: The diversified risk parity (DRP) strategy is used for multi-asset allocation to generate diversified equity portfolios. Creating uncorrelated risk sources by means of high-frequency principal components analysis (HF-PCA), we obtain maximum diversification portfolios when equally budgeting risk to each of the uncorrelated risk sources. We forecast the risk factors and trace the role of firms/industries as potential sources of financial risk in different periods of time. The empirical analysis carried out using one-minute returns of stocks included in the S\&P 100 index from 2003 to 2022 belonging to ten industry groups, shows that compared to classical risk-based allocation schemes, the DRP strategy provides the most convincing risk-adjusted performance and the most diversified portfolio among the investigated alternatives according to several concentration indices and risk decomposition characteristics. HF-PCA allows the DRP strategy to constantly adapt to changes in risk structure and maintain a balanced exposure to the prevailing uncorrelated risk sources. This tool can help a portfolio manager understand and choose risk sources that have earned risk by focusing on those risk factors.
