A Study Of The Effect Between Investor Sentiment And The Stock Return

In the traditional finance, investors are assumed to be rational person whosesentiments don’t have influences on stock returns. However, Changes of investors’sentiments would lead to stock prices’ deviating from their fundamental value as theresult of the investment being easily affected by factors like moods and emotions, aswell as the limited interest arbitrage in the market. Our country’s stock market is youngand there exists many deficiencies involving abnormal fluctuations of stock prices,obvious features of "policy market" and "informafion market", and unreasonableinvestors’ structure, which is far away from the level of effective market. Investors’sentiments have effects on the market that can’t be ignored. Therefore, this paper willdiscuss influences of investors’ sentiments on stock returns from the perspective ofbehavioral finance, which is a totally new vision to observe our country’s stock market.With regard to the study on investors’ sentiments, this paper firstly aims to solvethe problem of measurement towards investors’ sentiments. The text refers to the BWindex and chooses variables that could represent our country’s investors’ sentimentsfrom numerous ones on the consideration of our country’s practical situation, and thenbuilds the index of China’s investors’ sentiments after eliminating macroeconomicfactors with the use of principle cause analysis methods. This paper thinks thatinvestors’ sentiments influence stock returns through the systematic effect and crosssection effect on stocks of different features. Therefore, the paper studies the investors’sentiments influence on stock returns from two aspects. Firstly, with regard to the totaleffect, the text sets up relative regression model to empirically test the short-term andlong-term yield in the market. Take current stocks returns as the explained variable andmake regression analysis with current and phase1lagged sentiments index separately.The total effect of investors’ sentiments will be examined via judging the significance ofthe regression model. Take average stocks returns of future several months as theexplained variable and make regression analysis with current sentiments index. Thelong-term effect of investors’ sentiments will be examined via judging the significanceof the regression model in the same way. Secondly, regarding cross section effect ofinvestors’ sentiments, the paper sets up a dynamic portfolio to make out statisticallyaverage monthly returns of different companies’ stocks on the condition of differentsentiments, and analyzes the investors’ sentiments influence on stock portfolio ofdifferent features through the regression model. The results give indications as follows. Firstly, investors’ sentiments produce total effects on stock returns. In the short term,investors’ sentiments put positive influence on stock market returns; while in themedium and long term, investors’ sentiments still put positive influence on stock marketreturns which is contrary with most research conclusions in foreign countries. The thesisconsiders that it is due to the specific investors’ composition, market structure andmechanism. Secondly, investors’ sentiments have cross section effects on stock returns.After sequencing and grouping stock samples within the sample period according tocompanies’ features, using non-parametric statistics and regression analysis, we findthat investors’ sentiments have cross section effects on stock returns which stems fromstocks’ speculative difference and investors’ expectance uncertainty towards the future.Empirical tests figure out that the scale negatively correlates with investors’ sentimentalsensitiveness, with which financial exposure and corporate growth positively correlate,while corporate profitability does not have remarkable relations with investors’sentimental sensitiveness. Stocks of small value, high asset-liability ration and highprice-book ratio are prone to be affected by investors’ sentiments.Since behavioral finance is complex, this thesis is a shallow exploration forinvestors’ sentiments. Although this thesis relatively valuably studies investors’sentiments influence on stock returns from aspects of both the theory and empirical test,it still exists shortages in many respects. Firstly, for our country’s stock market is latelyestablished and investors’ sentiments indexes are not perfectly made, we can not useannual data which is mostly adopted in foreign study. So the thesis utilizes monthly datafrom January in2005to December in2011. This probably affects the reality ofconclusions. Secondly, the measurement of investors’ sentiments puzzles scholars’correspondent study. Investors’ sentiments are vague and elusive concepts. The indexbuilt in the thesis could reflect investors’ sentiments in the market, but can not totallyand comprehensively do. Thirdly, while observing investors’ sentiments cross sectioneffect on stock market, the thesis only adopts four variables of corporate features andthus can not make an overall reflection and research towards the degree of corporatestocks’ speculative.