股價漲跌是否引發網路熱搜反映投資人情緒？

Does The Rise and Fall of Stock Prices Trigger Hot Searches on The Internet to Reflect Investor Sentiment?

本文探討股價漲跌是否引發投資人情緒利用網路搜索資訊進而影響股票報酬，研
究採用Hansen（1999）提出縱橫門檻迴歸模型（Panel Threshold Regression，PTR），樣本涵蓋2012年7月至2022年6月10年期間臺灣82家資本額百億以上之上市公司週資料。研究發現股票報酬作為門檻變數得以區隔Google搜尋趨勢指數對股票報酬不同程度影響之門檻效果，投資人情緒對股票報酬影響於門檻-4.8086%上、下兩側形成區間分界，Google搜尋趨勢指數分別為正向顯著影響與負向顯著影響股票報酬，投資人易因搜尋資訊產生過度樂觀與過度悲觀反應，導致股票報酬持續上漲與下
跌，因此Google搜尋趨勢指數作為投資人情緒代理變數，得以解釋股票報酬，促進價格發現與市場效率，成為投資決策重要參考指標。

This article explores whether changes in stock prices affect investor sentiment and how they utilize internet search information to ultimately impact stock returns. The study employs Hansen’s (1999) panel threshold regression (PTR) model and uses weekly data from 82 listed companies in Taiwan with a capitalization of over NT$10 billion from July 2012 to June 2022 as the sample. The findings imply that utilizing stock returns as a threshold variable enables the differentiation of the threshold effects of Google search volume index, which have differing impacts on stock returns. Investor sentiment significantly impacts stock returns at a threshold of -4.8086%, which results in the formation of two interval divisions as Google search volume index has significant positive and negative effects on stock returns. Investors may overreact to digital information leading to prolonged increases or decreases in stock returns. Therefore, the Google search volume index should be regarded as a proxy variable for investor attention and used to explain stock returns to enhance price discovery and market efficiency, thereby becoming an important reference indicator for investment decision-making.