A hazard model of US airline passengers’ refund and exchange behavior

This study explores the use of discrete choice methods for airline passenger cancellation behavior. A discrete time proportional odds model with a prospective time scale is estimated based on the occurrence of cancellations (defined as refund and exchange events) in a sample of tickets provided by the Airline Reporting Corporation. Empirical results based on 2004 data from eight domestic US markets indicate that the intensity of the cancellation process is strongly influenced by both the time from ticket purchase and the time before flight departure. Higher cancellations rates are generally observed for recently purchased tickets, and for tickets whose associated flight departure dates are near. Cancellations rates are influenced by several other covariates, including departure day of week, market, and group size.