In line with the attention grabbing theory, the publication of articles dealing with the profile of single listed companies along with financial analysts' recommendations is followed by significant increases when the recommendation is positive. In this paper, we tried to understand what happens when analysts' recommendations are missing. We estimated analysts’ recommendations with a classification based on the terms in the articles and through a sentiment analysis of the same texts. Results showed that investors transform the articles’ content into implicit recommendations that guide their buying decisions when the sentiment is highly positive.
Media and Investors’ Attention. Estimating analysts’ ratings and sentiment of a financial column to predict abnormal returns / Ferretti, Riccardo; Sciandra, Andrea. - (2021), pp. 1543-1548. (Intervento presentato al convegno SIS 2021 - 50th Meeting of the Italian Statistical Society tenutosi a University of Pisa (virtual conference) nel 21-25 June 2021).
Media and Investors’ Attention. Estimating analysts’ ratings and sentiment of a financial column to predict abnormal returns.
Riccardo Ferretti;Andrea Sciandra
2021
Abstract
In line with the attention grabbing theory, the publication of articles dealing with the profile of single listed companies along with financial analysts' recommendations is followed by significant increases when the recommendation is positive. In this paper, we tried to understand what happens when analysts' recommendations are missing. We estimated analysts’ recommendations with a classification based on the terms in the articles and through a sentiment analysis of the same texts. Results showed that investors transform the articles’ content into implicit recommendations that guide their buying decisions when the sentiment is highly positive.File | Dimensione | Formato | |
---|---|---|---|
sis-book-2021-Ferretti-Sciandra.pdf
Open access
Tipologia:
Versione pubblicata dall'editore
Dimensione
854.46 kB
Formato
Adobe PDF
|
854.46 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris