Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/192899
Authors: 
De Nard, Gianluca
Ledoit, Olivier
Wolf, Michael
Year of Publication: 
2018
Series/Report no.: 
Working Paper 290
Abstract: 
This paper injects factor structure into the estimation of time-varying, large-dimensional covariance matrices of stock returns. Existing factor models struggle to model the covariance matrix of residuals in the presence of conditional heteroskedasticity in large universes. Conversely, rotation-equivariant estimators of large-dimensional time-varying covariance matrices forsake directional information embedded in market-wide risk factors. We introduce a new covariance matrix estimator that blends factor structure with conditional heteroskedasticity of residuals in large dimensions up to 1000 stocks. It displays superior all-around performance on historical data against a variety of state-of-the-art competitors, including static factor models, exogenous factor models, sparsity-based models, and structure-free dynamic models. This new estimator can be used to deliver more efficient portfolio selection and detection of anomalies in the cross-section of stock returns.
Subjects: 
Dynamic conditional correlations
factor models
multivariate GARCH
Markowitz portfolio selection
nonlinear shrinkage
JEL: 
C13
C58
G11
Persistent Identifier of the first edition: 
Document Type: 
Working Paper
Social Media Mentions:

Files in This Item:
File
Size
298.84 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.