| CAPM model plays a very important role in Finance Theory since the beginning. Its essential point is, at the time of equilibrium status of security market, the expected return on securities is equal to riskfree return plus a risk premium. That means, the expected return on securities are a positive linear function of their market βs. Unfortunately, the empirical record of the model is poor –poor enough to invalidate the way it is used in applications. The CAPM's empirical problems may reflect theoretical failings, the result of many simplifying assumptions. On the other side, the "Anomalies"has been gradually focused on. Variables like size, book-to-market equity and E/Pare all scaled versions of a firm's stock price. But the ability of βto interpret stock return is very poor. Such kind of demonstration rule contrary to classical CAMP Model is called "Anomalies". Currently there are various arguments and explanations concerning about such "Anomalies". Some economists represented by Roll think anomalies resulted from the fact that the real market portfolio can not be found. It would not be accurate to substitute market portfolio with some kind of stock index , which causes the failure of CAPM. Another arguments about anomalies is that anomalies is the result of data excavation, and is casual phenomenon. Behavior Finance Specialists regard anomalies as market ineffectiveness and unreasonableness of investors. On this aspect, the most important idea was raised by Lakonishok, Shleifer, Vishny. They argue that the size effect and value effect are caused by extra response of investors. This article analyzes the relationship between stock's average return and βvalue of stock, market equity(ME), book-to-market equity(BE/ME), earnings-price ratios(E/P) based on Shanghai stock market data from 1998 to 2004. It is shown out that βvalue, either as an independent variable or as one factor combined with others, can't interpret the fluctuation of stock return. But the other three variables, ME, BE/ME, E/P can well interpret the fluctuation of stock return. This article is divided into six chapters. The detailed structure is shown as following: Chapter one is preface and instruction of related literatures. Because capital evaluation is an essential question of Finance, related literatures are as many as smoke on the sea, we collected varies researches of scholars from domestic and abroad, among those theories, the research of Fama and French in 1992 is the most famous one, and is an comparatively comprehended research. It makes cross-sectional regression of the combined factors of βvalue, company size, book-to-market value and earnings-price ratio. It is the main theory frame we take as research reference for following chapters. Chapter two is the estimating of βvalue. The real amount of βvalue can only be assessed by sample instead of observation, therefore it is a key issue to estimate βvalue accurately and effectively. This article takes means of OLS to estimate βvalue. Considering the status quo that china's interest rate is not totally marketlised, and it is difficult to get risk-free interest rate, we use market model to estimate βvalue. In chapter three, we form portfolios on different scales, and give a direct observation how the β, ME, BE/ME, E/P affect the stock's average return. Chapter four and Chapter five make one-dimension and multiple-dimensions panel data regression of stock return and other variables. Using panel data model is one of the creations of this article; Formal researches mostly take the FM regression take cross-sectional data regressionanalysis, whose presumption is factors make same influence to different stocks. In the dimension of time sequence, FM regression just compute the equal-weighted average of regression slopes in each periods, it may smears some useful information on time sequence, and therefore, it has some limit. This article uses panel data model which take double dimensions of time sequence and section and can make clearer description of variable factors. combining this reality related to this article and making model distinction, we take variable-modulus fixed effect model as our empirical model and the slope from the regression is allowed to vary between different stocks. We get the main results as following: Market equity(ME) has strong power to explain the average stock returns. The negative relation between stock average return and size is prominent among every models. China's stock market shows strong "Size affect". Besides, in multiple-dimension regression, company's market equity is more capable to interpret as fixed modulus factor, it illuminates that the influence direction and extent of stock return from size factor is moving in the same way. βshows no power to explain the average returns in regression when βis the only variable. After adding variable of ME, and regarding it as modulus invariable factor, the prominence of βis improved to some extend, but the direction of regression modulus of βis uncertain, plus and minus share half to half. It illuminates that after separating the influence of size, βand stock return still do not show out the positive linearity relationship as expressed in CAPM model, but shows different trends among different stocks. BE/ME has positive relationship with stock average return. If the BE/ME is regarded as fixed modulus in regression, which means book-to-market...
|
PDF Full Text Request