Research On Investor Sentiment And Irrational Behavioral Biases In The Chinese Stock Market

Behavioral finance theory conbines psychology, sociology and economics, and it can explain the different kinds of anomalies in the stock market by studying the mental factors behand the actually investment behaviors. Investors always make mistakes because of their sentiment and mental factors, and behavioral finance helps us understand the importantance of mental factors in the stock market. This dissertation studies the investor sentiment and mental effects on investor behaviors biases in the Chinese stock market. The study on investors’sentiment and behavioral biases has important significance to the prediction of stock market bubbles and the policy formulation. The main research works of this dissertation are as follows:(l)Construct investor sentiment index. This dissertation constructs a proxy of investor sentiment for individual stocks and also figures out another investor sentiment measure which filters the market noise employing the Extended Kalman Filter (EKF) methodology. It lays a foundation for further empirical study on investor sentiment.(2) Empirical study on investor sentiment in the Chinese stock market. This dissertation employs the dynamic combination approaches and panel data models to test the relationships among investor sentiment, stock returns, stock capitalization and market attention. The results show that the volatility of stock returns increases with the investor sentiment. In addition, investor sentiment is comparatively lower for stocks with high market attention. The tests also show that investor sentiment for large-capitalization stocks is less volatile than that for the small-capitalization stocks. This empirical study presents evidence that investor sentiment plays an important role in Chinese stock market.(3) Study on the spillover effects between investor sentiment and stock returns. Using the multivariate GARCH models, this dissertation further investigates the relationship between investor sentiment and stock return volatility. After separating the emotion into rational and irrational sentiment, the results show that before the revolution, there was spillover effect existing between sentiment and stock returns while the rational and irrational sentiment had quite different effects on stock return volatility. And after the revolution, the relationship between investor sentiment and stock returns clearly waned. The strong relationship between investor sentiment and stock returns has practical meaning for comprehending the informational transfer and forecasting the stock return volatility.(4) Study on investor sentiment index and stock market based on extended Kalman Filter Method. By employing the Extended Kalman filter (EKF) methodology, this dissertation for the first time figures out the investor sentiment measure which filters the market noise. Further more, it investigates the investor sentiment and its relationships to stock returns and capitalization scales using the VAR model. The results show that a clear investor sentiment measure is extracted by the EKF methodology and the changes in sentiment has much more predictive power for the stock returns than the sentiment measure itself. We also find that the investor sentiment has much more effect on stock returns for small size companies than for the big size companies. Moreover, there exist a long-term negative and short-term positive influence on stock returns by sentiment volatility, which indicates that investor sentiment plays an important role in the asset price fluctuation.(5) Investors’behavioral biases and stop loss strategy:empirical study based on disposition effect and reference point. This dissertation studies the investors’ behavioral biases and their preference property in the Chinese stock market, and it confirms the reference price for investors. The results show that macroeconomic factors affect investors’preference. Before the split-share structure reform, investors preferred to ride losers too long, and sell winners too early. However, after the reform, investors prefer to hold the gains. On the other side, house money effect weakens the disposition effect. We also find that investors consider both dynamic costs and highest stock price as the reference point. While considering individual’s behavioral biases, the stop loss strategy, which helps the investors earn more money, is more predominant than the holding strategy. (6)Empirical study on reference point adaptation for fund managers in China. Based on the prospect theory, this dissertation for the first time studies empirically on the reference point adaptation for fund managers, and explores the relationships between reference point changes and stock market returns, shareholding costs, investor sentiment and stock market index. The results show that, fund managers raise their reference point slowly under profit, and lower their reference point quickly when they experience losses. We also find that the reference point fluctuation becomes stronger when fund managers are holding high-cost stocks compared to holding low-cost stocks. Investor sentiment is negatively related with reference point fluctuation. The study on the relationship between stock market index and reference point shows that fund managers are more sensitive to losses than to gains. Being influenced by the above factors, reference point adaptation will directly affect the fund managers’decision making and their investment behavior.At the end of the dissertation,it summarizes the results and contributions of the research, and points out the further research direction in the future.