Title: Singular spectrum analysis for value at risk evaluation in stochastic volatilty models
Authors: Josu Arteche - University of the Basque Country UPV/EHU (Spain) [presenting]
Javier Garcia - University of the Basque Country (Spain)
Abstract: Estimation of the Value at risk (VaR) in Stochastic Volatility (SV) models requires prediction of the future volatility, which is a function of a latent variable that is not observable. In-sample and out-of-sample prediction of that unobservable variable is thus necessary. The former is related with signal extraction whereas the latter implies prediction of future values. Singular Spectrum Analysis (SSA) is a useful tool for both purposes. The focus is mainly on out-of-sample predictions, comparing the performance of SSA with other forecasting techniques when used for VaR evaluation in SV models. Their empirical performance is also analysed in a daily series of SP500 returns.