Bivariate Stochastic Volatility Models Applied
 to Air Pollution Data

Jorge Alberto ACHCAR[1]

Henrique Ceretta ZOZOLOTTO1

Eliane R. RODRIGUES[2]

§    ABSTRACT: In this paper, we introduce the use of bivariate stochastic volatility models applied to air pollution data. Recent introduced stochastic volatility models used to analyze financial time series are considered to estimate the volatilities of the weekly ozone measures considering two different regions of Mexico city in the period from January 01, 1990 to December 31, 2005. The Bayesian analysis is developed using Markov Chain Monte Carlo (MCMC) methods to simulate samples for the joint posterior distribution of interest.

§    KEYWORDS: Stochastic volatility models; air pollution data; ozone pollution; Bayesian analysis; MCMC methods.



[1] Departamento Ciências Exatas, Escola Superior de Agricultura ``Luiz de Queiroz", Universidade de São Paulo, Caixa Postal 9, CEP 13418-900 , Piracicaba, São Paulo, Brasil. E-mail: rrsilva@esalq.usp.br /  sszocchi@esalq.usp.br

[2] Instituto de Matemáticas ­ UNAM, Area de la Investigacíon Científica Circuito Exterior, Ciudad Universitaria México, D.F. 04510, México. E-mail: eliane@math.unam.mx