Can Google data improve the forecasting performance of tourist arrivals? Mixed-data sampling approach

This paper introduces a new indicator for tourism demand forecasting constructed from Google Trends' search query time series data. The indicator is based on a composite search for “hotels and flights” from three main source countries to five popular tourist destinations in the Caribbean. We uniquely test the forecasting performance of the indicator using Autoregressive Mixed-Data Sampling (AR-MIDAS) models relative to the Seasonal Autoregressive Integrated Moving Average (SARIMA) and autoregressive (AR) approach. The twelve month forecasts reveal that AR-MIDAS outperformed the alternatives in most of the out-of-sample forecasting experiments. This suggests that Google Trends information offers significant benefits to forecasters, particularly in tourism. Hence, policymakers and business practitioners especially in the Caribbean can take advantage of the forecasting capability of Google search data for their planning purposes.