?
Применение Google Trends для прогнозирования миграции из России: агрегация поисковых запросов и учет лаговой структуры
This paper proposes a method for predicting migration based on search query data statistics using Google Trends Index (GTI). We improved the existing methodology in two directions: firstly, we proposed an approach for selecting key search queries and aggregating them based on various statistical criteria; secondly, we showed the importance of including in the migration model the time lag structure of search queries, depending on the goals of the migration and the associated GTIs. Based on the monthly data of the German statistical office on the volume of migration from Russia to Germany from January 2011 to August 2022, we demonstrate the efficiency of the proposed approaches. The results show that distributed lag migration models with GTI are better predict migration than SARIMA models. Average lag estimates, i.e. the response time of migration statistics to search queries on the topics “embassy”, “work” and “study”, turned out to be 5.6, 6.5 and 8 months, respectively. It is shown that in order to predict migration from Russia to Germany, it is sufficient to take into account only search queries related to the topic "embassy".