Template-type: ReDIF-Article 1.0
Author-Name: Alisa Polekhina
Author-Email: alisapolekhina@gmail.com
Author-Workplace-Name: Bank of Russia
Author-Name: Anna Guseva
Author-Email: adgusevazz@gmail.com
Author-Workplace-Name: Bank of Russia
Title: How the Bank of Russia Is Perceived on Telegram Channels: Building an Index Using Machine Learning Methods
Abstract: The paper constructs a Bank of Russia perception index on Telegram channels, which may serve as a leading indicator of public confidence in the regulator (correlation with InFOM survey data - 74%). The index is estimated on unstructured data from 1,400 Telegram channels. This is the first index of its kind, providing a comprehensive picture of the information field by classifying channels into types and key areas of the Bank of Russia's activities, from monetary policy to the financial market and the national payment system. For text analysis, we use both the traditional dictionary method and modern large linguistic models. The final index correlates with household inflation expectations and business price expectations but has no statistical link to financial market variables. The index opens up new opportunities for researching public perception of the Bank of Russia's policy and can be used as a tool for assessing the effectiveness of its communication.
Classification-JEL: C63, D83, E52, C55
Keywords: sentiment index, machine learning, natural language processing, semantic analysis, inflation expectations, Telegram
Journal: Russian Journal of Money and Finance
Pages: 28-62
Volume: 84
Issue: 3
Year: 2025
Month: September
DOI:
File-URL: https://rjmf.econs.online/upload/documents/RJMF-84-3-Telegram-Bank-Russia-Index.pdf
Handle: RePEc:bkr:journl:v:84:y:2025:i:3:p:28-62