Dynamic Conditional Correlation Models for Realized Covariance Matrices


Speaker


Abstract

New dynamic models for realized covariance matrices are proposed. The expected value of the realized covariance matrix is specified in two steps: a model for each realized variance, and a model for the realized correlation matrix. The realized correlation model is a dynamic conditional correlation model. Estimation can be done in two steps as well, and a QML interpretation is given to each step, by assuming a Wishart conditional distribution. Moreover, the model is applicable to large matrices since estimation can be done by the composite likelihood method.