| 摘要: | 在現代發達的科技社會中，網際網路普遍被使用。網路使用行為所留下的數據，例如使用量或對訊息的反應，會影響其他人的行為(Bordino, Battistion, Caldarelli, Cristelli, Ukkonen, and Weber, 2012)。現今人們可以透過網路搜尋取得更多相關資訊，並利用語意情緒分析，產出網路聲量。網路聲量代表網路上對特定標的之評價，而正的聲量代表投資人具有好的情緒；負的聲量則代表投資人具有負面情緒。情緒能深深影響個人的投資行為與決策(Bollen, Mao, and Zeng, 2011)。為探究網路聲量活動與股票市場之關係，本研究採用不對稱EGARCH模型，探討2015年07月30日至2016年07月26日間的網路聲量對台灣上市金融控股公司股票之影響。研究結果指出：(1)總聲量對報酬呈正向影響，對波動則呈負向影響；其中，正聲量對報酬與波動皆呈負向影響；負聲量對報酬呈負向影響，對波動則呈正向影響；(2)負聲量對股票市場之報酬與波動較正聲量敏感；(3)網路聲量對民營較泛公股之金融控股公司，對股票市場報酬與波動影響較大。本研究納入網路聲量，作為投資者之投資決策依據。
With the advancements of technology, the Internet is now widely used throughout the world. The behavior of Internet users, such as their usage or reactions to information, leave a digital trace and affect other people’s actions (Bordino et al., 2012). Nowadays, people can utilize Internet technology and surf the World Wide Web to obtain any kind of information, and sentiment analysis can be explored by Internet sentiment tracking. Internet sentiment tracking represents Internet evaluation on a specific target. Positive (negative) sentiment tracking represents positive (negative) sentiment. Emotions can profoundly affect individual behavior and decision-making (Bollen et al., 2011). Hence, EGARCH will be employed to investigate the internet sentiment tracking activities and stock returns and volatility of Taiwan’s listed financial holding companies from July 30, 2015 to July 26, 2016. The findings indicate that: (1)Total sentiments have a positive influence on returns, and negative influence on volatility; positive sentiments have negative influence on returns and volatility; negative sentiments have a negative influence on returns, and positive influence on volatility. (2) Negative sentiments are more sensitive than positive sentiments on stock returns and volatility. (3)Internet Sentiment had a greater effect on private financial holding companies than state-owned financial holding companies. The results of analysis provides investors with better reference information when making investment decisions. |
