Title: Robust principal volatility components
Authors: Carlos Trucios - Sao Paulo School of Economics - FGV (Brazil) [presenting]
Abstract: The recent principal volatility components procedure is analized. The procedure overcomes several difficulties in modelling and forecasting the conditional covariance matrix in large dimensions arising from the curse of dimensionality. We take into account the presence of outliers, which are common in financial time series, and show that outliers have a devastating effect on the construction of the principal volatility components and on the forecast conditional covariance matrix and consequently in economic and financial applications based of this forecast. To overcome this problem, we propose a robust procedure called robust principal volatility components and analyse its finite sample properties by means of Monte Carlo experiments. The procedure is also illustrated using empirical data. The robust procedure outperforms the non robust method in simulated and empirical data.