Forecasting Large Realized Covariance Matrices: The Benefits of Factor Models and Shrinkage
This work addresses the curse of dimensionality in financial forecasting for investors and analysts, though it appears incremental as it builds on existing factor and shrinkage methods.
The authors tackled the problem of forecasting large realized covariance matrices for S&P 500 returns by proposing a factor model with sectoral restrictions and LASSO-based VHAR estimation, resulting in improved forecasting precision and better minimum variance portfolio estimates.
We propose a model to forecast large realized covariance matrices of returns, applying it to the constituents of the S\&P 500 daily. To address the curse of dimensionality, we decompose the return covariance matrix using standard firm-level factors (e.g., size, value, and profitability) and use sectoral restrictions in the residual covariance matrix. This restricted model is then estimated using vector heterogeneous autoregressive (VHAR) models with the least absolute shrinkage and selection operator (LASSO). Our methodology improves forecasting precision relative to standard benchmarks and leads to better estimates of minimum variance portfolios.