Forecasting Large Realized Covariance Matrices: The Benefits of Factor Models and Shrinkage

50 Pages Posted: 11 May 2018

See all articles by Diego Brito

Diego Brito

Pontifical Catholic University of Rio de Janeiro (PUC-Rio) - Department of Economics

Marcelo C. Medeiros

Pontifical Catholic University of Rio de Janeiro (PUC-Rio) - Department of Economics

Ruy Ribeiro

Insper

Date Written: April 16, 2018

Abstract

We propose a model to forecast very large realized covariance matrices of returns, applying it to the constituents of the S&P 500 on a daily basis. 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 estimated with the least absolute shrinkage and selection operator (LASSO). Our methodology improves forecasting precision relative to standard benchmarks and leads to better estimates of the minimum variance portfolios.

Keywords: realized covariance, factor models, shrinkage, Lasso, forecasting, portfolio allocation, big data

JEL Classification: C22

Suggested Citation

Brito, Diego and Cunha Medeiros, Marcelo and Ribeiro, Ruy, Forecasting Large Realized Covariance Matrices: The Benefits of Factor Models and Shrinkage (April 16, 2018). Available at SSRN: https://ssrn.com/abstract=3163668 or http://dx.doi.org/10.2139/ssrn.3163668

Diego Brito

Pontifical Catholic University of Rio de Janeiro (PUC-Rio) - Department of Economics ( email )

Rua Marques de Sao Vicente, 225/206F
Rio de Janeiro, RJ 22453
Brazil

Marcelo Cunha Medeiros (Contact Author)

Pontifical Catholic University of Rio de Janeiro (PUC-Rio) - Department of Economics ( email )

Rua Marques de Sao Vicente, 225/206F
Rio de Janeiro, RJ 22453
Brazil
+55 21 3114-1078 (Phone)

Ruy Ribeiro

Insper ( email )

R Quata 300
Sao Paulo, 04542-030
Brazil

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
297
Abstract Views
1,175
Rank
155,904
PlumX Metrics