Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/309531 
Autor:innen: 
Erscheinungsjahr: 
2023
Quellenangabe: 
[Journal:] Credit and Capital Markets – Kredit und Kapital [ISSN:] 2199-1235 [Volume:] 56 [Issue:] 3-4 [Year:] 2023 [Pages:] 353-388
Verlag: 
Duncker & Humblot, Berlin
Zusammenfassung: 
Domain-specific dictionaries have prevailed, when conducting the dictionary-based approach to measure the sentiment of textual data in finance. Through the contributions of Bannier et al. (2019a) and Pöferlein (2021), two versions of a dictionary suitable for analyzing German finance-related texts are available (BPW dictionary). This paper conducts and tests further improvements of the given word lists by calculating the sentiment of German-speaking annual reports to forecast future return on assets and future return on equity. This corrected and expanded version provides more significant results. Despite the broad usage of negations, this type of improvement in combination with the BPW dictionary has not yet been tested when conducting the dictionary-based approach. Therefore, this paper additionally tests different negation lists to show that implementing negations can improve results.
Schlagwörter: 
Textual Analysis
Textual Sentiment
Sentiment Analysis
Content Analysis
Negations
Annual Reports
JEL: 
G14
G17
Persistent Identifier der Erstveröffentlichung: 
Creative-Commons-Lizenz: 
cc-by Logo
Dokumentart: 
Article

Datei(en):
Datei
Größe





Publikationen in EconStor sind urheberrechtlich geschützt.