Research Papers‎ > ‎

Realized Wishart-GARCH: A Score-driven Multi-Asset Volatility Model

posted Sep 1, 2016, 6:34 PM by Peter Hansen   [ updated Mar 18, 2019, 9:34 AM ]

Journal of Financial Econometrics Vol. 17 pp. 1-32 (2019) with Paolo Gorgi, Paweł Janus and Siem Jan Koopman.

We propose a novel multivariate GARCH model that incorporates realized measures for the covariance matrix of returns. The joint formulation of a multivariate dynamic model for outer-products of returns, realized variances and realized covariances. The updating of the variance matrix relies on the score function of the joint likelihood function based on Gaussian and Wishart densities. The dynamic model is parsimonious while the analysis relies on straightforward computations. In a Monte Carlo study , we show that parameters are estimated accurately for different small sample sizes. We illustrate the model with an empirical in-sample and out-of-sample analysis for a portfolio of 15 U.S. financial assets.


Peter Hansen,
Apr 20, 2018, 8:42 AM