Quantifying Sentiment for the Japanese Economy as Predictors of Stock Prices

Ishijima, Hiroshi

This paper quantifies market sentiment as four indexes and examines whether they can help predict
stock prices in Japanese markets. Sentiment analysis is gaining increasing interest in both academia
and business. Previously, Ishijima et al. (2014) created a sentiment index that quantifies the positive
or negative emotions that might appear in the Nikkei, which is the most popular business newspaper
in Japan. They concluded that the sentiment index significantly predicts stock prices three days in
advance. We re-examine their recent 5-year-worth results by extending in two dimensions; that is,
we extend the coverage of the Nikkei to 29 years and create variations of their original sentiment
index. On the basis of 29-year-worth daily sentiment indexes, we thoroughly examine the
predictability of Japanese stock prices. The findings of our year-by-year analysis are two-fold: (1)
sentiment indexes created from all of Nikkei’s articles persistently predict the Nikkei 225 stock prices,
in both in-sample and out-of-sample bases, and (2) these periods can be interpreted using business
cycles defined by Japan’s Cabinet Office.



  • thumnail for WP_338.Hiroshi_Ishijima.Quantifying_Sentiment.pdf WP_338.Hiroshi_Ishijima.Quantifying_Sentiment.pdf application/pdf 638 KB Download File

More About This Work

Academic Units
Center on Japanese Economy and Business
Center on Japanese Economy and Business, Graduate School of Business, Columbia University
Center on Japanese Economy and Business Working Papers, 338
Published Here
November 5, 2014