Zhang+05

テキスト分析

http://www.springerlink.com/content/f075n3834342g5m3/

Exchange Rate Modelling Using News Articles and Economic Data

書籍シリーズ Lecture Notes in Computer Science 出版社 Springer Berlin / Heidelberg ISSN 0302-9743 (Print) 1611-3349 (Online) 巻 Volume 3809/2005 書籍 AI 2005: Advances in Artificial Intelligence DOI 10.1007/11589990 著作権 2005 ISBN 978-3-540-30462-3 カテゴリー Decision Making DOI 10.1007/11589990_49 ページ 467-476

Debbie Zhang, Simeon J. Simoff and John Debenham (Faculty of Information Technology, University of Technology, Sydney)

Abstract This paper provides a framework of using news articles and economic data to model the exchange rate changes between Euro and US dollars. Many studies have conducted on the approach of regressing exchange rate movement using numerical data such as macroeconomic indicators. However, this approach is effective in studying the long term trend of the movement but not so accurate in short to middle term behaviour. Recent research suggests that the market daily movement is the result of the market reaction to the daily news. In this paper, it is proposed to use text mining methods to incorporate the daily economic news as well as economic and political events into the prediction model. While this type of news is not included in most of existing models due to its non-quantitative nature, it has important influence in short to middle terms of market behaviour. It is expected that this approach will lead to an exchange rate model with improved accuracy.