Modeling Passenger's Airline Ticket Choice Using Segment Specific Cross Nested Logit Model with Brand Loyalty

摘要

The proliferation of low cost carriers and the reduced cost of comparing alternatives have empowered airline passengers. Nowadays, it is easy for airline customers to compare products from numerous airlines and choose the one that suits them the best. Rapid growth in the number of similar products means that simplified methods of analyzing passenger behavior will give erroneous results. In this paper we present a novel methodology that extends the Cross-Nested Logit model. We then employ it to analyze airline choice for travel between Australia and the United States. The model is estimated using data from a stated preference choice experiment conducted among Australian citizens traveling to the United States. As a result of the specific design of the experiment we show that passenger purchase behavior depends heavily on the time the decision is made. In addition, supplementary questions have allowed us to propose a new segmentation model. The resulting discrete choice model renders almost identical results to the more complex Mixed Logit models, but without the additional computational burden.