Sensitivity of News Confidence Index to Taiwan Stock Market

• Main Authors: Yen-Ju Hsu, 許嫣茹
• Other Authors: Yu-Chen Wei
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/75896543912541657196

碩士 === 國立高雄第一科技大學 === 金融系碩士班金融組 === 101 === This study is based on a series research which focused on the relationship between public media information and stock market (Chen, 2003; Fang and Press, 2009; Engelberg and Parsons, 2011; Griffin, 2011). The linguistic analysis is employed to quantify the information content by extending the concept of linguistic analysis from Vega (2006) and Demers and Vega (2011). Further, this study constructs a market aggregate news confidence index (ANCI) by collecting the huge Chinese financial news related to each stock listed on the Taiwan Stock Exchange and considering the market value. The study period extends from January 2003 to December 2012 and employs individual stocks news of China times and Commercial times to construct the daily, weekly and monthly ANCI respectively. The findings show that trading volume, trading value, turnover ratio and volatility index will increase when the previous daily or weekly ANCI is positive. Furthermore, the investors’ fear gauge of the volatility index decreases when the lagged term of ANCI increases. The empirical results show that daily ANCI could be a leading index to the trading volume, trading value, turnover and volatility index in the Taiwan stock market. On the other hand, weekly and monthly ANCI could be a leading index to the market returns. The findings of portfolio application present that the lower(higher) confidence index, the higher the subsequent return, moreover, the result confirms that investors could buy selected portfolios with higher turnover, smaller size or lower stock price to book value while the previous higher confidence scenarios is distinguished. This study contributes to previous studies that the degree of confidence reflected by the qualitative news reports should be incorporated in the field of the sentiment proxy for decision making.