An Analysis of Airline Quality Rating Components Using Bayesian Methods

Abstract

Verbatim from Embry-Riddle Scholarly Commons. Not paraphrased, not summarized.

The Airline Quality Rating, a unique metric that is of interval scale and is comparable across carriers and time periods, is a quantitative evaluation of the quality of U.S. domestic air carriers based on data that is collected and published by the U.S. Bureau of Transportation Statistics. One may gain insight into the relationship of the AQR metric to the carriers to which it is applied by grouping those carriers and examining the effects of the resulting groupings on the four individual factors that comprise the AQR. This study used Bayesian hierarchical modeling techniques to examine the differences between three carrier groupings (legacy, low-cost, and regional) on a longitudinal basis over a period of six years. Based on the results of a Bayesian two-way analysis of variance (BANOVA), results showed significant differences between the mishandled baggage, denied boardings, and customer complaints of the data sets. Thus, sufficient evidence was found to support the premise that different econometric models are needed to broaden stride ins air carrier service quality improvements. Based on the results of the Bayesian two-way analysis of variance, credible differences in on-time arrivals, mishandled baggage, and customer complaints were indicated. These results imply a rejection of the null hypothesis, Ho, for those data sets. There were no credible differences indicated between legacy, regional, and low-cost carriers regarding the denied boarding data set, implying a failure to reject the null hypothesis in that particular case.

Authors

• Mott, John H. Embry-Riddle Aeronautical University
• Avery, Branden K Embry-Riddle Aeronautical University

Keywords

• Airline Quality Rating
• ANOVA
• Bayesian inference
• Econometric indicators
• Econometrics