Models of Artificial Neural Networks Applied to Demand Forecasting in Nonconsolidated Tourist Destinations
2015
This article focuses on a new proposed artificial neural network (ANN) model for tourism demand forecasting using time-series which, unlike previous models, uses different seasons of arrivals and values of months with similar behavior as input variables and achieves a forecast up to a year in advance. We demonstrate the validity and greater precision of the proposed model in forecasting a nonconsolidated destination with marked seasonality, which has been scarcely dealt with in other research. We achieve a comparatively greater quality of results and a longer period in advance than previously used auto-regressive integrated moving average (ARIMA) and ANN models. Highly accurate results were also obtained in destinations such as Portugal, which also proves its validity for mature destinations.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
98
References
7
Citations
NaN
KQI