Common Volatility in Evolutionary Panels

• Activity Number: 636
• Type: Invited
• Date/Time: Thursday, August 4, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Giovanni Motta*+ and Matteo Barigozzi
• Companies: Maastricht University and London School of Economics and Political Science
• Keywords: Approximate Factor Models ; Local Stationarity ; Asymptotic Principal Components

Abstract

Large panels of equity returns typically exhibit strong evidence of co-movement. In this work we consider a factor model for multivariate processes whose second order structure smoothly varies over time.

We factorize the evolutionary loadings as the product of a scalar smooth time-varying component that captures the long run common volatility, and ARMA filters describing the short run stationary dynamics. Specific non-pervasive behaviours of returns are left in the idiosyncratic components.

We then estimate the common volatility in a fully non-parametric way and derive its asymptotic properties. The performance of the methodology is illustrated by means of simulation exercises.

Finally, we provide an application to a panel of equity returns on the S&P 500 constituents. Empirical results show that there is a strong evidence of a factor structure and that, a large portion of the overall volatility is explained by the common volatility of the market.