【摘要】：有效市場假說(EMH)認為,一個有效的市場總能立即充分、準確地反映所有的信息。但是大量實證研究發(fā)現,現實的金融市場并不是這樣的,對違背有效市場假說(EMH)和傳統(tǒng)的資本資產定價模型(CAPM)的現象,我們稱其為“異�！�,這也正反映出傳統(tǒng)的方法在模型構建和估計方法上存在著問題。因此,合理分析“異�！爆F象,從而找到適合現實金融市場的資本資產定價方式是一個非常值得關注的重要研究方向。 本文針對出現“異常”現象的核心原因——模型的構建問題和傳統(tǒng)的回歸方法的缺陷進行了詳細分析,并由此提出了基于換手率的多因素模型,同時針對傳統(tǒng)回歸方法的缺陷,將分位數回歸方法運用到股票截面收益率的研究中。本文的創(chuàng)新之處如下： 首先,在Fama和French(1993)的三因素模型的基礎上,引入換手率作為一個新的解釋變量,得到基于換手率的多因素模型。通過針對我國深市A股的實證分析,發(fā)現換手率與截面收益率的相關性為正,而且相比其他解釋變量更加顯著。 其次,用Koenker和Bassett(1978)提出的分位數回歸方法代替?zhèn)鹘y(tǒng)的均值回歸方法,更適應現實的復雜金融市場,分位數回歸方法不僅可以研究在均值處風險因子對截面收益率的影響,而且還可以研究在收益率的條件分布上其它任何一點處風險因子對截面收益率的影響,這就使得研究更加全面完整。
[Abstract]:The efficient Market hypothesis (EMH) argues that an efficient market always reflects all information in a sufficient and accurate manner. However, a large number of empirical studies have found that this is not the case in real financial markets. We call this phenomenon "abnormal" when it goes against the efficient Market hypothesis (EMH) and the traditional capital asset pricing model (CAPM). This also reflects the problems of traditional methods in model building and estimation. Therefore, it is an important research direction to analyze the phenomenon of "anomaly" reasonably and find the capital asset pricing method suitable for the real financial market. In this paper, the core cause of "abnormal" phenomenon, the construction of the model and the defects of the traditional regression method, are analyzed in detail, and the multi-factor model based on the turnover rate is put forward, and the defects of the traditional regression method are also pointed out. The quantile regression method is applied to the research of stock cross section return. The innovations of this paper are as follows: firstly, based on the three-factor model of Fama and French (1993), the turnover rate is introduced as a new explanatory variable, and the multi-factor model based on the turnover rate is obtained. Based on the empirical analysis of A-shares in Shenzhen Stock Exchange, it is found that the correlation between turnover rate and cross-section return is positive, and it is more significant than other explanatory variables. Secondly, using the quantile regression method proposed by Koenker and Bassett (1978) to replace the traditional mean regression method, it is more suitable for the complex financial market. The quantile regression method can not only study the effect of risk factors at the mean value on the cross-section return. Moreover, we can study the effect of risk factors on the cross-section return rate at any other point in the conditional distribution of the return rate, which makes the research more comprehensive and complete.
【學位授予單位】：青島大學
【學位級別】：碩士
【學位授予年份】：2012
【分類號】：F832.51;F224

【引證文獻】

相關碩士學位論文 前1條

1 周鑫;資本資產定價模型及其擴展模型的實證比較研究[D];貴州財經大學;2013年

，

本文編號：2146840