CMStatistics 2017: Start Registration
View Submission - CFE
Title: Estimation of realized betas in a multi-factor model in presence of noise and asynchronisity Authors:  Sebastien Laurent - AMU (France) [presenting]
Orimar Sauri - Aarhus University (Mexico)
Silvana Acosta - Aarhus (Denmark)
Abstract: An estimator of realized betas in a multi-factor model is proposed which is robust to microstructure-noise and asynchronisity. The estimator relies on the Cholesky factorization of the realized covariance matrix and, therefore, it also provides positive semidefinite estimates of the covariance matrix. We prove consistency and asymptotic normality of the estimator of the realized betas. Monte Carlo simulations confirm good finite sample properties. We apply our estimator on 52 stocks US stocks and 10 ETFs.