Asset Pricing Using Block-Cholesky GARCH and Time-Varying Betas

Starting from the Cholesky-GARCH model, recently proposed by Darolles, Francq, and Laurent (2018), paper introduces Block-Cholesky GARCH (BC-GARCH). This new model adapts in a natural way to asset pricing framework. After deriving conditions for stationarity, uniform invertibility beta tracking, we investigate finite sample properties of variety maximum likelihood estimators suited BC-GARCH means an extensive Monte Carlo experiment. Finally, illustrate usefulness two empirical applications. The first tests presence spillovers bivariate system context Fama French (1993) three factor second application consists large scale exercise exploring cross-sectional variation expected returns 40 industry portfolios.