A Multi-Model Time Series Framework for Forecasting Vietnam’s Tourism Revenue in the Post-COVID Recovery Era

Ho Nhat Hiep (Faculty of Commerce, Van Lang University, Ho Chi Minh City, Vietnam)
Nguyen Ngoc Xuan Quynh (Faculty of Commerce, Van Lang University, Ho Chi Minh City, Vietnam)
Nguyen Thi Van Anh (Faculty of Commerce, Van Lang University, Ho Chi Minh City, Vietnam)
Minh Ly Duc (Faculty of Commerce, Van Lang University, Ho Chi Minh City, Vietnam)

Sinkron : Jurnal dan Penelitian Teknik Informatika
Vol. 10 No. 3 (2026): Article Research July 2026

This study forecasts Vietnam’s tourism revenue in the post-COVID-19 recovery period using a multi-model time series framework. The dataset includes three groups (ID 292–294) covering tourism business performance, economic sectors, and regional revenue. Six forecasting models are applied and evaluated using MAPE, MAD, and MSD. Results show that decomposition and Holt–Winters achieve the best accuracy (e.g., MAPE as low as 18%), while moving average performs well in specific cases (MAPE ≈ 28%). Forecasts indicate that tourism revenue may nearly double by 2030, driven mainly by domestic demand and the non-state sector, although international tourism recovers more slowly.