Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/267319 
Year of Publication: 
2022
Series/Report no.: 
CESifo Working Paper No. 10087
Publisher: 
Center for Economic Studies and ifo Institute (CESifo), Munich
Abstract: 
The availability of copious amounts of data produced by the increasing datification of our society is nowadays deemed an opportunity to produce timely and convenient statistical information. This paper shows the building of economic sentiment indexes from the texts of the most read economic newspapers in Spain. The data are collected through the scraping of the Digital Periodical and Newspaper Library website. To compute the sentiment, an existing emotional lexicon for Spanish words has been customized, allowing for inferring sentiment for words in texts. The resulting indexes are later compared to other well-known indicators that try to monitor similar or related phenomena.
Subjects: 
index numbers
large datasets
leading indicators
proxy variables
sentiment analysis
web scraping
JEL: 
C18
C43
C55
C89
Document Type: 
Working Paper
Appears in Collections:

Files in This Item:
File
Size





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