An ARIMAX model incorporating Google search trends effectively forecasts monthly tourist arrivals from India to Nepal, aiding policymakers and stakeholders in decision-making.
Authors
Biplab Bhattacharjee, Associate Professor, Jindal Global Business School, O.P Jindal Global University, Sonipat, Haryana, India
Aayush Poudel, Kathmandu School of Management, Kathmandu University, Nepal, India
Subin Panta, Kathmandu School of Management, Kathmandu University, Nepal, India
Sashwat Sharma, Kathmandu School of Management, Kathmandu University, Nepal, India
Samyak Pokharel, Kathmandu School of Management, Kathmandu University, Nepal, India
Summary
Tourism is a crucial component of the economy in developing countries like Nepal. Nepal is a popular tourist destination for India due to its proximity, shared cultural heritage, and affinities between the people and their respective religions. This study attempts to develop an econometric forecasting model to predict the monthly arrivals from India to Nepal. The COVID-19 pandemic significantly impacted the travel and tourism industry, causing a structural break in the time series arrival data. Using monthly data from 2004 to 2034, the study applies time series models to address complexities such as seasonality, non-stationarity, and structural breaks (due to COVID-19). The findings reveal that an ARIMAX model incorporating Google search trends data performs better than traditional models based on several evaluative measures such as RMSE, MAPE, AIC, and Theil’s U. The proposed forecasting model can assist policymakers, hotel management, and event planners in estimating the level of tourism demand and making better managerial decisions.
Published in: Springer Proceedings in Business and Economics
To read the full article, please click here.
