Modeling competition among airline itineraries

Discrete choice models are commonly used to forecast the probability an airline passenger chooses a specific itinerary. In a prior study, we estimated an itinerary choice model based on a multinomial logit specification that corrected for price endogeneity. In this paper, we extend the analysis to include inter-itinerary competition along three dimensions: nonstop versus connecting level of service, carrier, and time of day using nested logit (NL) and ordered generalized extreme value (OGEV) models. To the best of our knowledge, these are the first NL and OGEV itinerary choice models to correct for price endogeneity. Despite the many structural changes that have occurred in the airline industry, our results are strikingly similar to models estimated more than a decade ago. These results are important because it suggests that customer preferences, on average, have been stable over time and are similar across distribution channels. The stability in inter-itinerary competition patterns provides an important practical implication for airlines, namely it reduces the need to frequently update the parameter estimates for these models.

[1]  Frank S. Koppelman,et al.  Schedule delay impacts on air-travel itinerary demand , 2008 .

[2]  M. Ben-Akiva,et al.  Endogeneity in Residential Location Choice Models , 2006 .

[3]  H. Williams.  On the Formation of Travel Demand Models and Economic Evaluation Measures of User Benefit , 1977 .

[4]  Gregory M Coldren.  Modeling the competitive dynamic among air-travel itineraries with generalize extreme value models , 2005 .

[5]  Frank S. Koppelman,et al.  Modeling the Proximate Covariance Property of Air Travel Itineraries Along the Time-of-Day Dimension , 2005 .

[6]  D. Rivers,et al.  Limited Information Estimators and Exogeneity Tests for Simultaneous Probit Models , 1988 .

[7]  M. Bierlaire,et al.  Choice Probability Generating Functions , 2012 .

[8]  F. Koppelman,et al.  The generalized nested logit model , 2001 .

[9]  K. Small.  A Discrete Choice Model for Ordered Alternatives , 1987 .

[10]  Laurie A. Garrow,et al.  Discrete Choice Modelling and Air Travel Demand: Theory and Applications , 2010 .

[11]  Virginie Lurkin,et al.  Accounting for Price Endogeneity in Airline Itinerary Choice Models: An Application to Continental U.S. Markets , 2016 .

[12]  Virginie Lurkin,et al.  Accounting for Price Endogeneity in Airline Itinerary Choice Models: An Application to Continental U.S. Markets , 2016 .

[13]  Frank S. Koppelman,et al.  Modeling the competition among air-travel itinerary shares: GEV model development , 2005 .

[14]  Cynthia Barnhart,et al.  Quantitative problem solving methods in the airline industry : a modeling methodology handbook , 2012 .

[15]  C. Angelo Guevara,et al.  Critical assessment of five methods to correct for endogeneity in discrete-choice models , 2015 .

[16]  Frank S. Koppelman,et al.  Modeling aggregate air-travel itinerary shares: logit model development at a major US airline , 2003 .

[17]  Daniel McFadden,et al.  Modelling the Choice of Residential Location , 1977 .

[18]  Virginie Lurkin,et al.  Computational methods for estimating multinomial, nested, and cross-nested logit models that account for semi-aggregate data , 2018 .

[19]  J. M. Villas-Boas,et al.  Endogeneity in Brand Choice Models , 1999 .

[20]  D. McFadden.  Conditional logit analysis of qualitative choice behavior , 1972 .

[21]  Gregory K. Leonard,et al.  Competitive Analysis with Differentiated Products , 1994 .

[22]  J. Stock,et al.  Instrumental Variables Regression with Weak Instruments , 1994 .

[23]  Frank S. Koppelman,et al.  Airline Planning and Schedule Development , 2012 .