Abstract
This paper estimates the volatility of most important
European stock market indices during the global financial
crisis started in 2008, such as DAX, CAC40, FTSE100,
among others. The estimation of volatility is made from a new family of stochastic
volatility models proposed by Santos, Franco, Gamerman [33, 17] and extended to
distributions of heavy tails by Pinho, Franco, Silva [32]. This new
family of models denoted by non-Gaussian State Space Models
(NGSSM) is a subclass of state space models where it is possible to
compute the exact likelihood. It is also estimated volatility of the
series by APARCH model and the results showed that NGSSM has a
significantly better performance.