This paper examines the behavior of the US airline industry's service deman
d. Monthly aggregate data for the industry are analyzed. While we find stro
ng evidence of nonlinear dependence in the air transport service time serie
s, the evidence is not consistent with chaos. We also show that GARCH model
s successfully explain the nonlinear structures in the US airline industry'
s service series. Finally, within-sample forecasts of air transport demand
from the GARCH models outperform those of simple autoregressive models. (C)
2001 Elsevier Science Ltd. All rights reserved.