Identifying common trends in tourism data

The aim of this thesis is to identify common trends in the development of the number of overnight stays in tourism in nine European countries – Germany, Switzerland, Austria, Slovenia, Liechtenstein, the Czech Republic, Poland, Slovakia and Hungary – and to assess the influence of key macroeconomic factors, including real GDP per capita, the unemployment rate and the financial stress index, on the dynamics of these time series. The analysis focuses on both the absolute values of the number of overnight stays and their short-term changes expressed using first differences, and uses two dynamic factor analysis models: a variant without the inclusion of macroeconomic covariates and a model extended by the mentioned covariates. Results based on non-standardized data and a different specification of the covariance matrix of the error component are also presented in the margins. Dynamic factor analysis models allow the identification of latent factors describing the common dynamics of the development of the number of overnight stays across the monitored countries and the evaluation of factor loadings that show the degree of connection of individual countries to the common trend. The inclusion of macroeconomic covariates allows for a better understanding of the extent to which these variables influence the shape of the latent factors and the overall model fit. The models are estimated using the Kalman Expectation–Maximization and Broyden–Fletcher–Goldfarb–Shann methods, and the paper further compares different specifications of the covariance error structure. To complement this, the results are also interpreted in relation to short-term fluctuations and differences between individual countries. The obtained latent variables capture in particular the recovery after 2009, the stabilization before 2019, the significant decline caused by the COVID-19 pandemic, and the subsequent return of tourism activity. At the same time, the results reveal specific dynamics, especially in Switzerland and Liechtenstein, which deviate from the main regional trend due to the structural conditions of tourism itself.