Understanding the factors affecting airport level demand (arrivals and departures) using a novel modeling approach

Abstract

The current study proposes a novel modeling approach for modeling airline demand. Specifically, we develop a joint panel generalized ordered probit model system with observed thresholds for modeling air passenger arrivals and departures while accommodating for the influence of observed and unobserved effects on airline demand across multiple time periods. The proposed model is estimated using airline data compiled by Bureau of Transportation Statistics for 510 airports in the US at a quarterly level for five annual time points. A host of independent variables including demographic characteristics, built environment characteristics, spatial and temporal factors are considered. From the model estimation, the important factors affecting airline demand include metropolitan statistical area (MSA) population, median income, education attainment, airport location and temporal factors. A validation exercise is also performed using a holdout sample to highlight the superior performance of the proposed model. Finally, to illustrate how the proposed demand model allows agencies to understand changes to airline demand with changes to independent variables, a policy analysis is conducted.