Fractional integration and fat tails for realized covariance kernels

We introduce a new fractionally integrated model for covariance matrix dynamics based on the long-memory behavior of daily realized covariance matrix kernels. We account for fat tails in the data by an appropriate distributional assumption. The covariance matrix dynamics are formulated as a numerically efficient matrix recursion that ensures positive definiteness under simple parameter constraints. Using intraday stock data over the period 2001-2012, we construct realized covariance kernels and show that the new fractionally integrated model statistically and economically outperforms recent alternatives such as the multivariate HEAVY model and the multivariate HAR model. In addition, the long-memory behavior is more important during non-crisis periods.