Tourism flow forecasting for inbound European travel

• Autor(a) principal: Garcia, André Guilherme Ramalho
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/30719
• topic: European tourism; Tourism forecasting; Forecasting models; Turismo Europeu; Previsão de turismo; Modelos de previsão

Tourism plays a pivotal role in the European economy. Among other aspects, the management of demand of products and services is critical to tourism development, so forecasting models contribute to such endeavour. This is even more relevant nowadays as the sector is recovering from the disruption caused by the global COVID-19 pandemic. In this study we build state-of-art forecasting models for tourism demand. In particular, models based on Deep Learning algorithms, such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), and considering both daily and monthly data frequency. But the contribution goes beyond mere algorithm selection – it also delves into the nuances of feature engineering, incorporating data from exogenous variables such as search engine volume of searches, inflation, GDP and currency exchange rate. Through rigorous evaluation, supported by forecasting evaluation metrics, such as RMSE, R-squared and MAPE, it was discovered that GRU models consistently outperformed LSTM models. Additionally, our exploration revealed that the inclusion of external factors had limited impact on enhancing forecast accuracy. This work serves as a valuable resource for industry stakeholders beyond the academic realm. Its findings are used to deploy forecasts into a web-application developed in the context of the European Union funded project RESETTING, which aims to help SMEs of the tourism sector as they make a comeback from the tough restrictive years during the pandemic.