Investor sentiments and stock market behaviour in the UK : an empirical analysis

Abstract

In finance literature, the behaviour of stocks is often analogised with that of a drunkard because of the erratic and unpredictable movements in prices. Consequently, over the years, stock traders have devised strategies to predict and possibly outperform the market, even though it is generally believed that stock markets cannot be beaten consistently and persistently. Similarly, several academic researchers have proposed theories, approaches and concepts, aimed at exploring trading strategies and patterns. A recent stratum that has been widely embraced in modern finance is the role of human sentiment on the behaviour of stocks and its subsequent impact on market outcomes. This aspect of finance, commonly referred to as behavioural finance, emphasises the irrationality and subjectivity of investors in decision making. This thesis focuses on the impact of investor sentiment on the behaviour of UK stock market. By analysing the behaviour, this study specifically examines the effects of activities of sentiment-induced investors on both risk and returns in the UK stock market. Hence, this thesis is divided into three empirical chapters. The first empirical chapter examines the relationship between sentiment-apt investors and UK stock returns, at sector level, using monthly data, from January 1988 to December 2017. Adopting two new sentiment proxies (laggards to leaders and growth opportunity index) and analysed using self-exciting threshold autoregressive (SETAR) model, the chapter provides novel evidence on how sectoral returns in the UK stock market react to the activities of sentiment-prone investors. The findings reveal that aggregate returns in the sector are affected by activities of investors who embark on profit-taking when there is an increase in the proportion of lagging to leading stocks beyond the threshold value. Furthermore, when using the growth opportunity sentiment proxy, the chapter reports that the increase in growth above the growth threshold value has a significant impact on sectoral returns. The second empirical chapter explores the inclusion of sentiment proxies as a risk factor in asset pricing. Specifically, the chapter investigates whether investor sentiment indices improve the performances of extant asset pricing models - CAPM, 3 Factor models and 4 Factor models. Using UK monthly data from January 1993 to March 2017, we observe that the sentiment-induced models produce a small distance error compared to the traditional models, thus validating the use of sentiment measures in our asset pricing mechanism. In the first two empirical chapters, the impact of activities of sentiment-disposed investors on stock market returns was established. In the third chapter, new insights into the impact of human psychology on market volatility are documented. Using UK daily market data, from January 1988 to December 2017, the third chapter empirically outlines the roles of lagged volatility, asymmetric shocks, and investors’ sentiments on the behaviour of stocks within the UK financial sector. With the aid of a variety of generalised autoregressive conditional heteroscedasticity (GARCH) models, the model specifications are classified into the mean equation (returns) and variance equation (volatility). The chapter documents that sentiment and volatility levels prior to the financial crisis were not as high as during and post the crisis. Essentially, based on the information selection criteria, the findings reveal that volatility in the sector is significantly influenced by its own lagged volatility, unanticipated news, events, and shocks. More importantly, the study also shows that the sentiment variable significantly predicts volatility, thus indicating that volatility in the sector at any given period is essentially characterised by preceding investors’ sentiment in the sector. The findings are robust having been subjected to different model and residual diagnostic tests. This thesis offers some important recommendations to the academic community, stock market participants and regulators. It shows that sentiment is a crucial systematic risk factor and by implication, all stakeholders must take serious cognisance of its propensities when formulating policies and procedures for the stock market. We show that four factor model is inadequate and needs a sentiment factor to improve asset pricing. This study is informative to academia by revealing that traditional asset pricing models may not be sufficient to capture anomalies in asset pricing. We show that conditioning on sentiment is important to predict volatility of assets. The findings from this study can be useful to investors on market timing and in designing the appropriate investment and trading strategies for their stocks or portfolio. Our results also provide instructive measures to regulators on how to checkmate the activities of market participants, especially those, whose actions distort stock market standards. We also recommend transparency of the market and education of investors to reduce sentiment-based decisions. This thesis therefore confirms that sentiments have significant implications on the behaviour of stock markets.