||Forecasting extreme seasonal tourism demand
Fukushige, MototsuguElamin, Niematallah
Discussion Papers In Economics And Business
31 , 2016-09 , Graduate School of Economics and Osaka School of International Public Policy (OSIPP) Osaka University
When we face an extreme seasonal demand such that off-season monthly demands go down almost to zero, it is important to forecast the high-season demand only. In forecasting tourism demand for a place like a cold-water island, we meet such an extreme seasonality. However, in such places, tourism is one of the most important industries, and to forecast tourism demand in the high season is important. Rishiri Island in Japan is a typical example. We construct some forecasting models and evaluate their forecasting performance using several criteria. Forecasts from the SARIMA model with full-season data and without a trend term perform best in postsample evaluation. This shows that off-season data also provide useful information for forecasting the high-season tourism demand.