Abstract
The Tourism sector is of strategic importance to the North Region of Portugal and is growing. Forecasting monthly overnight stays in this region is, therefore, a relevant problem. In this paper, we analyze data more recent than those considered in previous studies and use them to develop and compare several forecasting models and methods. We conclude that the best results are achieved by models based on a non-parametric approach not considered so far for these data, the singular spectrum analysis.
Original language | English |
---|---|
Pages (from-to) | 2927-2940 |
Number of pages | 14 |
Journal | Journal of Applied Statistics |
DOIs | |
Publication status | Published - 2020 |
Bibliographical note
Publisher Copyright:© 2020, © 2020 Informa UK Limited, trading as Taylor & Francis Group.
Keywords
- Forecasting
- neural networks
- overnight stays
- singular spectrum analysis
- time series