Working Paper: NBER ID: w8554
Authors: Robert F. Engle; Kevin Sheppard
Abstract: In this paper, we develop the theoretical and empirical properties of a new class of multi-variate GARCH models capable of estimating large time-varying covariance matrices, Dynamic Conditional Correlation Multivariate GARCH. We show that the problem of multivariate conditional variance estimation can be simplified by estimating univariate GARCH models for each asset, and then, using transformed residuals resulting from the first stage, estimating a conditional correlation estimator. The standard errors for the first stage parameters remain consistent, and only the standard errors for the correlation parameters need be modified. We use the model to estimate the conditional covariance of up to 100 assets using S&P 500 Sector Indices and Dow Jones Industrial Average stocks, and conduct specification tests of the estimator using an industry standard benchmark for volatility models. This new estimator demonstrates very strong performance especially considering ease of implementation of the estimator.
Keywords: Multivariate GARCH; Dynamic Conditional Correlation; Time-Varying Covariance
JEL Codes: C32; G0; G1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| DCC multivariate GARCH model (C22) | estimation of time-varying covariance matrices (C51) |
| univariate GARCH models (C29) | estimation of conditional covariance matrices for up to 100 assets (C10) |
| transformed residuals (C51) | estimation of time-varying correlation matrix (C10) |
| DCC model (C59) | accuracy of covariance estimation (C51) |
| DCC model (C59) | implications for portfolio management and volatility modeling (C58) |