Forecasting Japanese tourism demand in Taiwan using an intervention analysis

The objective of this research is to assess whether two events, the 9‐21 Earthquake in 1999 and the Severe Acute Respiratory Syndrome outbreak in 2003, had a temporary or long‐term impact on the inbound tourism demand from Japan. Furthermore, a comparative study is conducted to assess whether intervention analysis produces better forecasts compared with forecasts without intervention analysis.

The data adopted in this study consist of monthly visitor arrivals from Japan to Taiwan for the period January 1979‐September 2006. The first 321 observations ( January 1979‐September 2005) are used to develop two tentative models, with and without intervention analyses, and then compare with the known values (October 2005‐September 2006) for accuracy testing.

Experimental results show that the effect of both disasters on Japanese inbound tourism presented only temporarily, and the forecasting efficiency of ARIMA with intervention is superior to that of a model without intervention.

The study had difficulty accurately delineating the rebound in Japanese tourist based on monthly data. There are other factors that might influence a rebound, such as people' fading memories or the purpose of visitation. The geographic proximity of Taiwan to Japan could also account for perceived risk factors.

The results indicate that the Japanese inbound arrivals sharply dropped following both of the two disastrous occurrences, suggesting that the Japanese tourists are likely to be responsive to prompt marketing strategies and messages. The practical implication for tourism operators include the usefulness of reinforcing the package holiday by establishing an attractively priced travel package or offering a package with a variety of highly desirable or unique features to increase competition.

This study is a first attempt in the tourism literature to model Japanese demand for travel to Taiwan after these two traumatic crises.