Quantifying news reports to proxy "Other Information" in ERC models

Kuo Tay Chen, Jian Shuen Lian, Yu Ting Hsieh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Many previous studies have investigated how earning announcement affects stock price. They measure the effect by employing earning response coefficient (ERC) models. However, the traditional models did not explicitly consider textual information received by investors. Rather they simply referred to it as ″other information″. However, investoŕs exposure to textual information (e.g. news report) might have significant influence on how stock prices will respond to earning announcements. This study attempts to investigate whether earning surprises cause stock fluctuations and how the effect is influenced by news coverage prior to earning announcements. We find that: (1) earning surprise significantly affects stock price; (2) more news coverage tends to decrease the ERC; (3) positive earning surprises have higher influence on stock price; and (4) different combinations of news sentiment and earning surprise result in different ERC.

Original languageEnglish
Title of host publicationIntelligence and Security Informatics - Pacific Asia Workshop, PAISI 2009, Proceedings
Pages161-168
Number of pages8
DOIs
Publication statusPublished - 2009
EventPacific Asia Workshop on Intelligence and Security Informatics, PAISI 2009 - Bangkok, Thailand
Duration: 2009 Apr 272009 Apr 27

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5477
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherPacific Asia Workshop on Intelligence and Security Informatics, PAISI 2009
Country/TerritoryThailand
CityBangkok
Period09-04-2709-04-27

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Quantifying news reports to proxy "Other Information" in ERC models'. Together they form a unique fingerprint.

Cite this