Modeling tourism demand is essential to the planning of this activity by those responsible for tourism policies in each region. This fact has led to the development and testing of different methodologies and its comparison with the goal of finding the “best” model. Comparison of different methods not clearly conclude on the best way to model the tourism demand and the majority of studies is concerned with finding a model that allows making good forecasts in the short and medium term. The tourism industry is very susceptible to specific events, so it is important not only to find good forecasting models, but also to study the volatility of this industry over time. This article aims to provide a systematic review of the recent literature targeting the models used to analyze this volatility. The recent literature reveals some determinants of volatility of the tourism industry, such as currency devaluation, the absence/existence of direct flights, climate change, economic crises, events and shocks among others. Moreover, this study reveals that the main approaches used in the analysis of time series volatility include generalized autoregressive conditional heteroscedasticity models, Markov chain models, grey forecasting models, exponential smoothing models, and neural networks, among others. This paper offers avenues for future research and discusses the managerial implications for decision makers.

--

Tourism demand,Volatility,Modelling,Tourism economics,Search engine data