Long/Short Equity Risk Premia Parity Portfolios via Implicit Factors in Regularized Covariance Regression

A time series basis decomposition and trend extraction technique known as Empirical Mode Decomposition (EMD), designed for multiscale time-frequency decomposition in nonstationary time series settings, will be combined with Regularised Covariance Regression (RCR) methods to produce a novel framework: the EMD-RCR covariance forecasting model. This will produce a model framework capable of generating multi-time resolution adaptive forecasting models of predictive covariance forecasts for a universe of selected asset returns. This provides a unique method to obtain predictive covariance regression structures for the study of short- and long-time-scale portfolio dynamics. The forecast time resolution is controlled by the time resolution of the extracted EMD factors.<br><br>An illustration is developed for active portfolio asset management, based on a dynamic risk-parity portfolio-of-portfolios investment strategy. In illustration of this methodology we use the eleven sector-based portfolios, sector exchange traded funds (ETFs), from the S&amp;P500 which are termed in this work sector indices. From each of these sector indices one can construct dynamically evolving equal risk parity portfolio framework utilising the EMD-RCR methodology developed. The portfolio will be reweighted monthly based on the covariance structure forecast using covariance regression, in which covariance regression factors will be obtained at multiple time-frequency scales endogenously from the ETF asset price returns from each sector. The performance of the portfolios will be measured using multiple performance measures and contrasted against multiple benchmark portfolios using several well-known portfolio optimisation techniques.<br><br>Empirical Mode Decomposition (EMD), Singular Spectrum Analysis (SSA), and Singular Spectrum Decomposition (SSD) will be used to isolate different frequency structures in the price data to be used as covariates in covariance regression to optimise a risk parity portfolio with weighting restrictions. This paper serves to promote the use of what we term “implicit factor" extraction and RCR in the interrelated fields of portfolio optimisation, horizon-specific active portfolio optimisation, long/short equity portfolios, and risk parity portfolios.