Forecasting Thailand inbound tourist flow association for tourism demand

Abstract

The purpose of this paper is to study the tourism demand of Thailand from six major source countries (CN, JP, KO, MAS, SG and LA) in two regions of Asia (East Asia and Southeast Asia). The data is from 1997q1 to 2020q1, which can also provide some advice for the recovery of Thailand tourism post COVID-19. Compared with previous studies, we consider the dependence of tourist flow in the same region. Precisely, we estimated the impact of tourism flows between pairwise countries in each region. This also provides a complement to the subsequent research on the structure of tourism dependence for tourism demand of Thailand. The Copula-based approaches are increasingly being used for tourism demand, we apply the copula-ARDL (ECM) framework to forecast tourist arrivals from two regions. Using the characteristics of correlation matrix in the trivariate Gaussian copula model, we have innovatively performed unconstrained parametric optimization of these two models to ensure the accuracy of the forecasting.