The objective of this study was to estimate and forecast the hourly ozone concentration in the Region of Grande Vitória, Espírito Santo, Brazil, using the ARMAX-GARCH model, for the period from 2011/01/01 to 2011/12/31. Data set from the State Environmental Institute (IEMA) was used. The models has run for three local stations: Laranjeiras, Enseadá do Suá and Cariacica. Some parameters measured at the stations were taken as explanatory variables of ozone concentration. These variables significantly improved the model estimated. The hourly forecasts for 2011/12/31 (chosen to verify the model accuracy) were very close to the observed values and the estimated ones generally followed the path of daily ozone concentration. When compared with the ARMA and ARMAX models, ARMAX-GARCH model proved to be more effective in the prediction of ozone pollution episodes (hourly concentration higher than 80 µg/m3), reduction on the number of false alarms and lowering the rate of undetected episodes.
ozone; air pollution; time series, ARMAX; GARCH
