本研究屬於市場微結構範疇，並參考Lee and Ready(1991)的情緒建構演算法- tick test建立了日內情緒指標BSI作為本研究的主要變數，以探討情緒變數與台灣個股報酬之間的關係。相較於以往使用Panel Data利用因子模型做為研究方法，本研究採用了VAR模型，以符合情緒指標和個股報酬資料的時間序列特性。同時，本研究引入馬可夫轉換模型，針對不同產業以及2020年新冠疫情前後進行了分析，以研究情緒變數在不同市場狀態下的不對稱性，其中將市場狀態分為牛市以及熊市兩種。研究選取了2017年至2022年的全資料區間以及2020年新冠疫情前後，共三個樣本區間進行討論。結果顯示，在整個資料區間中，情緒變數在牛市條件下相較於熊市條件對個股報酬具有更好的預測能力。而根據不同產業的分析結果顯示，情緒變數在不同狀態和不同產業中存在著不對稱性。另外，對比2020年新冠疫情前後的分析結果，本研究證明了在新冠疫情爆發後，情緒因子對台灣股市的影響結構發生了轉變，情緒變數對個股報酬的影響，在疫情後效果顯著下降，且不對稱特性不再明顯。

This study belongs to the field of market microstructure and refers to the emotion construction algorithm - tick test developed by Lee and Ready (1991) to establish the intraday sentiment indicator BSI as the main research variable. The study aims to explore the relationship between emotional variables and individual stock returns in Taiwan. In contrast to previous studies that used panel data, this research adopts a VAR model to capture the time series characteristics of the sentiment indicator and stock returns data. Additionally, a Markov switching model is employed to analyze different industries and the periods before and after the COVID-19 outbreak in 2020, in order to investigate the asymmetry of emotional variables under different states.

The study selects three sample periods for discussion: the entire data interval from 2017 to 2022, as well as the periods before and after the COVID-19 outbreak in 2020. The results indicate that, in the entire data interval, the sentiment variable exhibits better predictive ability for stock returns under bullish market conditions compared to bearish market conditions. Furthermore, the analysis of different industries shows that there is asymmetry in emotional variables across different states and industries. When comparing the results before and after the COVID-19 outbreak in 2020, this study demonstrates a structural change in the impact of sentiment factors on the Taiwanese stock market. After the COVID-19 pandemic, the influence of emotional variables on stock returns significantly decreased, and the asymmetry characteristics became less prominent.

第一章 緒論----------------------1
第一節 前言---------------------1
第二節 研究動機 -----------------2
第三節 研究目的 -----------------3
第四節 研究架構 -----------------5
第二章 文獻回顧 -----------------5
第一節 情緒指標建構--------------6
第二節 馬可夫轉換模型------------12
第三章 研究方法-------------------15
第一節 研究假說------------------15
第二節 MSVAR模型設定-------------16
第三節 羅吉斯迴歸模型建立---------25
第四章 資料描述及敘述性統計--------27
第五章 實證結果分析----------------29
第一節 MSVAR模型結果分析----------29
第二節 影響情緒變數顯著性之因子----35
第六章 結論------------------------38
第七章 參考文獻--------------------40

[1] 吳靖東, 2014.投資人情緒對股票報酬之影響-馬可夫狀態轉換模式之應用. Journal of Innovation and Management Vol.10 No.4
[2] 王毓敏、陳碧秀、殷壽鏞、林家妃, 2008.台灣股票市場波動率結構性變動與長期記憶性質, 國立虎尾科技大學學報，27卷3期
[3] Chung, S., Hung, C., Yeh, C., 2010. When does investor sentiment predict stock returns? Journal of Empirical Finance, 217-240.
[4] Fama, E.F., & French, K.R., 1993. Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33(1), 3–56.
[5] Gao, B., & Liu, X., 2020. Intraday sentiment and market returns. International Review of Economics and Finance, 69, 48–62.
[6] Huang, C., Yang, X., Yang, X., Sheng, H., 2014. An Empirical Study of the Effect of Investor Sentiment on Returns of Different Industries, Mathematical Problems in Engineering,1-11
[7] Kim, K., & Ryu, D., 2021. Term structure of sentiment effect on investor trading behavior. Finance Research Letters, 102005.
[8] Koy, A., & Akkaya, M., 2017. The Role of Consumer Confidence as a Leading Indicator on Stock Returns: A Markov Switching Approach. Annals of the University Dunarea de Jos of Galati: Fascicle: I, Economics & Applied Informatics, 23(1), 36-47
[9] Lee, C.M., & Ready, M.J., 1991. Inferring trade direction from intraday data, The Journal of Finance 46(2), 733–746.
[10] Li, J., 2021. The term structure effects of individual stock investor sentiment on excess returns, International Journal of Finance & Economics 26, 1695– 1705.
[11] Nooijen, S., J., & Broda, S., A., 2016. Predicting Equity Markets with Digital Online Media Sentiment: Evidence from Markov-switching Models. Journal of Behavioral Finance,321-33
[12] Renault, T., 2017. Intraday online investor sentiment and return patterns in the U.S. stock market, Journal of Banking & Finance, 25-40
[13] Schmeling, M., 2009. Investor sentiment and stock returns: Some international evidence, Journal of Empirical Finance 16(3), 394-408.
[14] Tetlock, P.C., 2007. Giving content to investor sentiment: the role of media in the stock market, The Journal of Finance 62, 1139–1168
[15] Verma, R., Soydemir, G.,2009. The Impact of individual and institutional investor sentiment on the market price of risk. The Quarterly Review of Economics and Finance, 49(3), 1129–1145.
[16] Yang, C., & Zhou, L., 2015. Investor trading behavior, investor sentiment and asset prices. The North American Journal of Economics and Finance, 42-62
[17] Yang, C., and Gao, B., 2014. The term structure of sentiment effect in stock index futures market, The North American Journal of Economics and Finance 30, 171–182.