Analysis of International Air Passenger Flows between Two Countries in the APEC Region Using Non-parametric Regression Tree Models

Gravity models have been widely applied to analyze the various cross border flows and economic activities between cities or countries. To estimate a gravity model, (parametric) linear regression techniques have been commonly employed to develop the relationship between passenger flows and factors that can significantly influence the flows. However, parametric regression models have their own model assumptions and pre-defined underlying relationship between dependent and independent variables. If these assumptions are violated, the model could lead to erroneous estimation of the flows between countries. Classification and Regression Tree (CART), one of widely applied data mining techniques, has been commonly applied in business administration, industry, and engineering. CART does not require any pre-defined underlying relationship between target (dependent) variable and predictors (independent variables) and have been shown to be powerful tools particularly in dealing with prediction and classification problems. This study collected data of 2006 and 2007 international air passenger flows between countries in the APEC region. A parametric linear regression and a non-parametric regression tree models were developed to establish the empirical relationship between air passenger flows and multiple factors, including distance, population, GPD, average income, unemployed rate, and many other economy-related variables. The estimation results from the linear regression model and the CART model are similar in general. Both models show that GDP, unemployment rate, import/export, distance and many other factors are the key determinants of international air passenger flows between two countries. By comparing the empirical findings between the linear regression and regression tree models, this study demonstrates that non-parametric regression tree models are good alternative methods for analyzing cross-country air passenger flows.