本研究旨在探討比特幣報酬率與市場情緒指數之間的關聯性，研究期間涵蓋2017年至2023年，資料頻率為週資料。研究方法採用最小平方法（OLS）進行迴歸分析，並納入五項情緒指數作為自變數，分別為：代表美國散戶投資人情緒的美國個人投資人情緒指數（AAII）、代表傳統避險資產情緒的CBSI黃金情緒指數、反映整體市場情緒的CBSI股票市場情緒指數、代表美元貨幣情緒的CBSI美元指數情緒指數，以及納斯達克綜合指數用以觀察比特幣報酬率與傳統科技股市之互動關係。 本研究主要發現如下： 1. 實證結果顯示，CBSI股票市場情緒指數的變動對比特幣報酬率具有顯著正向影響，顯示專業機構對股市的情緒變化能有效預測比特幣的價格走勢。相對地，AAII與CBSI黃金情緒指數對比特幣報酬率的影響則不具顯著性。 2. 研究亦發現，專業機構對美元指數的情緒變化與比特幣報酬率呈現負向關係，顯示美元情緒與比特幣之間存在替代性資產的特性。 3. CBSI黃金情緒指數與比特幣報酬率之間缺乏顯著關聯，顯示比特幣與傳統避險資產之間存在脫鉤現象。 綜上所述，本研究有助於深化對不同投資人情緒，特別是專業機構情緒，對比特幣報酬率波動影響的理解，並補充比特幣市場與情緒指數互動關係之相關文獻。 ;This study aims to explore the relationship between Bitcoin returns and various market sentiment indices, using weekly data spanning from 2017 to 2023. The analysis is conducted through Ordinary Least Squares (OLS) regression, incorporating five sentiment indices as independent variables: the American Association of Individual Investors (AAII) Sentiment Index, representing U.S. retail investor sentiment; the CBSI Gold Sentiment Index, reflecting sentiment toward traditional safe-haven assets; the CBSI Equity Market Sentiment Index, capturing institutional sentiment toward the stock market; the CBSI U.S. Dollar Sentiment Index, indicating sentiment toward the U.S. dollar; and the Nasdaq Composite Index, which serves as a proxy to examine the interaction between Bitcoin returns and the traditional technology equity market.
The empirical findings are summarized as follows: 1. Changes in the CBSI Equity Market Sentiment Index have a significantly positive impact on Bitcoin returns, suggesting that institutional sentiment toward the stock market possesses predictive power over Bitcoin price movements. In contrast, sentiment changes in the AAII and CBSI Gold Sentiment Indexes show limited statistical significance. 2. Institutional sentiment toward the U.S. dollar exhibits a negative relationship with Bitcoin returns, indicating a potential substitution effect between Bitcoin and the dollar. 3. The CBSI Gold Sentiment Index shows no significant correlation with Bitcoin returns, highlighting a decoupling between Bitcoin and traditional safe-haven assets.
Overall, this study contributes to a deeper understanding of how different investor sentiments, particularly those of institutional participants, influence the volatility and dynamics of Bitcoin returns, and enriches the literature on the interaction between cryptocurrency markets and sentiment indicators.
