Three essays on investor sentiment in financial markets

<p dir="ltr">This thesis comprises three distinct, but interrelated studies on the roles of investor sentiment in the financial markets. Previous studies show that traditional finance theories have historically fallen short of fully explaining financial market movements. The rise of the behavioural finance literature in recent decades further confirms the significant role of investor sentiment in explaining various puzzles and anomalies in financial markets. This thesis contributes to this literature by offering both systematic reviews and empirical evidence on the perspective of investor sentiment.</p><p dir="ltr">The first study (Chapter 2) of this thesis provides a bibliometric analysis and systematic literature review on the topic of investor sentiment during the past three decades. This chapter identifies the established and emerging themes of the investor sentiment literature by utilising the scientific and replicable approach to analysing 1,274 high-quality publications from 1991 to 2023. Three main richer arrays of topics include the effects of sentiment on asset prices, market risks and the cross-sectional of asset returns. In addition, four emerging themes are also revealed: sentiment and crises, sentiment and alternative investments, sentiment and social finance, and innovative sentiment measures. From the starting point of this study, each identified theme has significant potential for a new stream of research. This is the first study that systematically explores the sentiment literature from its initial development, providing a valuable research agenda for the next two chapters of this thesis.</p><p dir="ltr">The second study (Chapter 3) addresses one of the emerging themes identified in Chapter 2—investor sentiment and social finance, which is studied by exploring the impact of the driving forces of cross-market divergences on investor sentiment. This study utilises a global dataset of 52 stock markets and employs 40 country-specific factors in nine different clusters: (i) Financial development; (ii) Market structure; (iii) Cultural backgrounds; (iv) Religion backgrounds; (v) Legal origin; (vi) Market integrity; (vii) Institutional quality; (viii) Educational backgrounds; and (ix) Gambling opportunity. First, this study confirms that sentiment, in general, negatively predicts stock returns and positively predicts volatility at the global level. Second, we find that sentiment has a more persistent influence in more developed markets, whereas its effects are more immediate in emerging and frontier markets. Notably, sentiment's impact is more pronounced in countries with lower institutional holdings, greater limits to arbitrage, more gambling opportunities, lower internet availability, and more prone to herd-like culture and overreactions. Additionally, variations in the predictive power of sentiment are positively associated with literacy rates, common legal systems, institutional integrity, institutional quality, market integrity, and the prevalence of Catholic, Islamic and Buddhist populations. Therefore, this study is the first to provide strong evidence on how economic, political, cultural, and social factors can modify the impacts of investor sentiment on predicting the stock market movement, regarding both return and volatility.</p><p dir="ltr">The third study (Chapter 4) fills a gap in the existing literature by examining the potential influence of investor sentiment in a non-traditional asset class – the cryptocurrency market. This study investigates the role of investor sentiment in shaping the reactions of the cryptocurrency market to macroeconomic announcements. Using a substantial intraday dataset on the top 100 cryptocurrencies from 2014 to 2021, we demonstrate that the cryptocurrency market reacts significantly, in terms of returns and trading volume, to the announcements of the U.S macroeconomic news and other major economies, such as the European Union, China, Germany, and Japan. Leveraging the cryptocurrency sentiment indices provided by Thomson Reuters MarketPysch Indices (TRMI), this study confirms that sentiment plays a crucial role in modifying the reactions of crypto investors. During periods of bullish sentiment, the market's responses are notably subdued, showing a 40 to 50 percent reduction compared to bearish sentiment periods. These effects persist across various sub-sample analyses, alternative sentiment measures, estimation techniques, and additional controls for financial market and economic uncertainties. This study is the first to validate the impact of investor sentiment on cryptocurrency market reactions to macroeconomic announcements, providing valuable insights for market participants and regulators.</p><p dir="ltr">In a broader sense, this thesis's findings propose several distinct contributions to several overlooked but important roles of investor sentiment in the financial markets. As such, this thesis can contribute some strokes to the gigantic but incomplete picture of the field of behavioural finance.</p>