Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/322058 
Year of Publication: 
2025
Series/Report no.: 
ECB Working Paper No. 3047
Publisher: 
European Central Bank (ECB), Frankfurt a. M.
Abstract: 
Word embeddings are vectors of real numbers associated with words, designed to capture semantic and syntactic similarity between the words in a corpus of text. We estimate the word embeddings of the European Central Bank's introductory statements at monetary policy press conferences by using a simple natural language processing model (Word2Vec), only based on the information and model parameters available as of each press conference. We show that a measure based on such embeddings contributes to improve core inflation forecasts multiple quarters ahead. Other common textual analysis techniques, such as dictionary-based metrics or sentiment metrics do not obtain the same results. The information contained in the embeddings remains valuable for out-of-sample forecasting even after controlling for the central bank inflation forecasts, which are an important input for the introductory statements
Subjects: 
Embeddings
forecasting
central bank texts
Inflation
central bank
economic forecasting
monetary policy
natural language processing
JEL: 
E31
E37
E58
Persistent Identifier of the first edition: 
ISBN: 
978-92-899-7216-1
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.