Tourism demand nowcasting using a LASSO-MIDAS model

Purpose This study aims to examine whether and when real-time updated online search engine data such as the daily Baidu Index can be useful for improving accuracy of tourism demand nowcasting once monthly official statistical data, including historical visitor arrival macroeconomic variables, become available. Design/methodology/approach is first attempt use LASSO-MIDAS model proposed by Marsilli (2014) field forecasting deal with inconsistency in frequency curse problem caused high dimensionality data. Findings The empirical results context arrivals Hong Kong show that application a combination produces more accurate MIDAS-type models are used. effectiveness indicates penalty-based MIDAS option using high-dimensional mixed-frequency Originality/value represents progressively compare there any differences between aforementioned two types different frequencies nowcast demand. also contributes literature presenting evaluate applicability nowcasting.