Modelling Thailand tourism demand: A dual generalized maximum entropy estimator for panel data regression models

Abstract

© 2017 by the Mathematical Association of Thailand. All rights reserved. This study examines the factors that influence the behavior of international tourists to Thailand by using a dual generalized maximum entropy estimator for panel data regression models. The advantage of the entropy approach is its capability to deal with ill-prosed problem and the entropy approach for panel data has not yet been investigated in the tourism literature. The focus is on the tourists from 10 countries of origin having the highest number of international tourist arrivals to Thailand including Laos, Malaysia, Singapore, China, Japan, Korea, Russia, United Kingdom, USA, and India over the period of 22 years (1995−2016). A number of important economic factors, income, price, exchange rate, and number of population, are studied regarding international tourism demand. The study compares the results of two methods, namely ordinary least squared estimator and generalized maximum entropy estimator. According to minimum value of mean square error, the generalized maximum entropy estimator perform better than the ordinary least squared. The results of tourism demand estimation show that the growth in income of Thailands major tourists originating countries, exchange rate, and number of population in countries of origin have positive impact on international visitor arrivals to Thailand while relative price has a negative impact on international visitor arrivals to Thailand. The study also finds that per capita national income enjoy strong predictive power for Thailand tourism demand.