CLASSIFYING NEWS ANNOUNCEMENTS USING NAÏVE BAYES METHOD TO PREDICT EURO / DOLLAR VOLATILITIES
No Thumbnail Available
Files
Date
2008
Journal Title
Journal ISSN
Volume Title
Publisher
Universitas Widyatama
Abstract
Commonly used analysis of price movements in foreign exchange (forex) market are fundamental analysis and technical analysis. One among many indicators which influences the forex price is news articles. In this study, it was selected and classified news announcements which affected euro/dollar return volatilities. By using the Naïve Bayes theorem, the news was “weighted” to become the predictor of the forex price movements. The post-announcement reactions were highlighted and analyzed. They were classified and labeled as: “up”, “down”, or “unchanged”. The study revealed a significant predictive power of news announcements over the forex price movements of euro/dollar return volatilities.
Description
Keywords
Naïve Bayes, forex, news announcement