Özet:

The prediction and modeling work that will be carried out using scientific-based techniques in the tourism sector is an important source for the development programmes that will be prepared by the central and local government managers and the planning work of the representatives of the tourism sector. In Turkey, the number of work in which tourism revenues are modeled, predicted and forecasts for the future are produced is limited. The aim of this study, which is taken out of the vacuum seen in literature, is to model the external active tourism revenues of Turkey with the methodology of Box-Jenkins, the artificial nerve networks (YSA) and the methods of superior correction according to the data characteristics, with the help of the model achieved the high forecast success, the production of monthly tourism income forecasts for 2020. The estimated performance of the models is assessed by MAPE statistics. As a result of the analysis, the model that has been applied has been identified as the YSA model with the architecture [5:7:1], with the help of this model the monthly tourism income forecasts for 2020 have been produced.