The airline industry is susceptible to exogenous events and environmental changes. Particularly, the global financial crisis and the threat of the H1N1 influenza pandemic during 2008-2009 have presented considerable challenges to the air travel industry. The motive of this study is therefore to evaluate the capabilities of short-term air passenger volume estimation in Taiwan by employing grey system theory. A comparison of the accuracy of the four forecasting models, namely GM (1, 1), GM (1, 1) rolling model, Autoregressive (AR) and Exponential smoothing, with regard to predictive power in air passenger movements, is conducted. Experimental results indicate that the forecasting efficiency of GM (1, 1) rolling model is superior to other forecast methods. The results can offer valuable insights and provide a basis for further research in model building for short-term estimation under the circumstances of significant market fluctuations, data incompleteness, or information insufficiency.