TOURIST DEMAND FORECASTING MODELS

Abstract

The article analyzes models of tourist demand forecasting. The complexity of the tourism industry, combined with the influence of various economic, social, political, environmental and cultural factors, creates considerable difficulties for accurate forecasting of tourist demand. Current forecasting models may not give sufficiently correct conclusions due to the complex interaction of these factors. In addition, the unpredictable consequences of crises, such as economic downs, political instability, epidemic and other negative effects, emphasize the need to create strong and adaptive forecasting models. Research on tourist demand forecasting models is an important task that is crucial for sustainable development and successful competition in tourism. The relevance of this topic is determined by the needs of the industry in accurate and intended strategies aimed at meeting the needs of the modern tourist. Tourism uses different approaches to forecasting, which can be divided into quantitative, based on mathematical calculations and qualitative methods, which, in turn, can be different types of expert assessments. Works devoted to the analysis of tourism development using a quantitative approach can be divided into two main groups: causal (econometric) models and inaccessible (time series) tools. Often -used time rows are ARIMA, GARCH models, as well as econometric models (such as CI), Error Correction Models (ECM), Time Models (TVP), Vector AutoRegression (VAR). An alternative solution is the use of qualitative forecasting methods that make it possible to take into account many defining factors, including those that are not available in any mathematical equation, such as the likelihood of terrorism or natural catastrophes, a change in public opinion. In order to make the right choice of the method of forecasting in the field of tourism, it is necessary to take into account the available resources (economic, practical, human) and their possible influence on the chosen methods of forecasting at the stage of their choice. In addition, it is recommended to focus on existing data on the advantages and disadvantages of specific techniques, as well as to consider the basic requirements for the creation of models.