Recognizing Korea Discount: Analysis of the Relationship between News Sentiment and Stock Prices during 1994-2012

• In Event: 239 - Section on Communication and Information Technologies Roundtable Session (one-hour).
  In Refereed Roundtable: Table 06. News, Opinions, Framing and Content

Abstract

The purpose of this paper is to examine whether there is a statistical correlation between the tone of the press coverage and the stock volatility of the Korean firms. We applied Machine Learning method to collecting data, learning classification, and classifying news sentiment. We collected 4,484 articles of five newspapers (New York Times, Financial Times, The Wall Street Journal, Washington Post, and The Times) that contained the keyword "North Korea" in their headline for recent 20 years. Based on the sentiment of articles we classified automatically, we conducted simple regression analysis to test the relationship between news sentiment and stock price of Korean firms on the New York Stock Exchange (NYSE). We consistently found out that while the number of articles of the day did not influence the stock prices, the number of the previous day influenced stock prices of the day significantly. The results indicate that investors’ recognitions are constructed through negative / positive tones of the press, and the recognition influenced investment behavior.

Authors

• Yoonjung Park, Yonsei University

• Youngjin Kim, Yonsei University

• Min Song, Yonsei University

• Yonghak Kim, Yonsei University