Time Series Analysis of COVID-19’s Impact on the European’s Hospitality Industry: Estimates Based on Accommodation Statistics from 18 Countries

In this study we quantitatively analyze adverse effects of the COVID-19 pandemic-related restrictions on a hospitality industry’s demand indicator dynamics in 18 European countries. We propose a unified approach for measuring COVID-related losses based on comparing actual outcomes with counterfactual outcomes obtained using the state-of-the-art Prophet procedure. We show that the algorithm is very competitive compared to some of the widely used automated forecasting procedures. Using data disaggregated to segments based on countries, accommodation types, tourists’ residence, and month, it has been shown that losses are heterogeneous across various segments of the hospitality market. The relative importance analysis using a decomposition of explained variance in regression models has shown that most of the explained variance can be attributed to time effects followed by country effects. The role of tourists’ residence and accommodation type is much smaller, but still significant. The most and the least impacted groups of countries are identified.