seminario

[Pedro Morettin <pam@ime.usp.br>]

Pedro Morettin <pam@ime.usp.br>

Seminário

Séries Temporais, Análise de Dependência e
Aplicações em Atuária e Finanças.

Dia 25/10/2005, Terça Feira, 17 horas, Sala 247, Bloco A.


   How Persistent is Stock Return Volatility? An Answer with Markov 
       Regime Switching Stochastic Volatility Models 


                     Pedro L. Valls Pereira 
                         Ibmec São Paulo

Abstract
We use stochastic volatility models with Markov regime changing state equation 
to  investigate the important properties of volatility in stock returns, high 
persistence and smoothness. We show that volatility is far less persistent and 
smooth than the GARCH or  stochastic volatility models suggest. 
Our further analysis shows that volatility is less persistent in high 
volatility regimes than in  low volatility regimes. These results indicate 
that the extreme persistence we frequently  observe from daily stock index 
returns is led by changes in regime as well as persistent  low volatility 
regimes.


----- End forwarded message -----

Pedro Morettin <pam@ime.usp.br>