Forecasting Consumer Service Prices During the Coronavirus Pandemic Using Neural Networks: The Case of Transportation, Accommodation and Food Service Sections Across E.U

This study examines how the coronavirus pandemic may affect the price of consumer services in the Transportation, Accommodation and Food Service sections in the European Union over the next period utilizing Machine Learning. For the purpose of the study, the authors use monthly reports of coronavirus cases and deaths along with a nominal sample size of 44.000 units, mainly national institutes, from the Joint Harmonized EU Programme of Business and Consumer Surveys by Directorate-General for Economic and Financial Affairs of European Commission. The dataset contains balanced answers from surveys asking for positive and negative replies measuring managers’ assessment of their company’s turnover from past experience and future estimations. The authors present evidence that it is possible to forecast future expectations on service price evolution during the pandemic utilizing Neural Network models. These models can predict a balanced percentage which can further be used for a systematic decision-making process. This percentage depends on the number of cases and deaths in each country but not in the same analogy to others. Each country performs differently in every sub-category of economic activity presented. To the best of our knowledge, this is a first attempt to investigate and predict the impact of coronavirus on consumer service price. These predictions concern the evolution of economic indicators using Neural Networks. In case of emergency situations, such as during pandemic, it is difficult to have enough data to make reliable predictions using other statistical models, therefore utilizing machine learning methods seems appropriate.