Data Analytics For Forecasting Arrival of Tourism Visit in Indonesia

Abstract

Previous research has proven that online search data can be used to estimate tourist visits. Massive query data is a challenge in determining the right keywords to used to build indexes. This study proposes a framework for building a composite search index. The proposed forecasting framework emphasizes keyword selection using keywords in previous research and is based on the tourism organization's website. Previous studies predict the number of tourist arrivals at a tourist place, city or city-state, whereas in this study predict foreign tourist visits at the country level, Indonesia. Using the econometric model, this study uses data on Foreign Tourist Visits (FTV), Google Trends Index (GTI), Customer Price Index (CPI) and Exchange Rate (ER). The forecasting method uses the Vector Error Correction Model (VECM) and then analyzes the model, provides forecasting and structural analysis of the model. The results of the model analysis are long-term and short-term analyzes and the results of forecasting evaluation show that the MAPE score is good.