【作者】 李金凤；

【导师】 陈国进；

【作者基本信息】 厦门大学 ， 金融学， 2014， 硕士

【副题名】基于CAPM、Fama-French模型和HAM

【摘要】 2009年10月23日,有中国纳斯达克之称的创业板在深圳证券交易所正式开市。创业板市场是对我国主板市场的重要补充,亦是我国资本市场的重要组成部分,但国内对创业板市场的实证研究仍比较空白。学习和研究西方资产定价理论体系,探索适用于我国创业板市场的科学定价理论,揭示影响创业板市场股票收益率的风险因素和定价机制,为理性的投资者提供科学的决策依据,对促进我资本市场的健康发展具有重要的理论和实践意义。本文拟从理论模型分析和实证检验两方面综合探讨市场风险、账面市值比、公司规模、调整成本和公司绩效等风险因素对资产价格的影响,尝试寻找影响创业板市场股票定价的真正因素。在CAPM Fama-French三因素模型和基于习惯形成和调整成本的定价模型(Pricing Model Based on Habit Formation and Adjust Cost, HAM)理论模型的基础上,本文采用深圳创业板市场个股周收益率数据进行实证研究,系统分析了影响创业板市场股票收益率的风险因素和各资产定价模型在创业板市场的适用性。对CAPM的实证结果发现：标准的CAPM模型拟合度极差,且并未在创业板市场上检测到模型所揭示的贝塔值与股票收益之间的正相关关系。对三因素模型的实证分析发现：Fama-French三因素模型拟合程度较好,市场因子、规模因子、价值因子三个因素对目标组合收益率的波动有较好的解释力,其中市场因子的解释力最强,价值因子的解释力次之,规模因子相对最弱。同时,我们发现在样本区间内,创业板市场存在“小公司效应”和“价值效应”,即小规模公司股票收益率高于大规模公司,价值股收益率高于成长股。对HAM的实证分析发现：模型拟合程度优于Fama-French三因素模型和CAPM,四个定价因子对股票收益率的解释力最强的为市场因子,调整成本因子次之,其次是价值因子,而绩效因子对股票收益率波动的影响最小。更多还原

【Abstract】 The Growth Enterprise Market (GEM) was launched on October23,2009. As an important complement to Main-Board market, the GEM is being heralded as an important step forward for China’s stock and capital markets. However, there is still few empirical study of the growth enterprise market in domestic. There is all-important significance in both theory and practice for studying on Asset Pricing Theory. This article uses several foreign theoretical models and research findings to study The Growth Enterprise Market in order to establish a practical and effective pricing model in GEM, which helps investors to make decision.This dissertation makes a systematic analysis and summarization about the Capital Asset Pricing Model, Fama-French Three-Factor Model and the Pricing Model Based on Habit Formation and Adjust Cost. Based on the theoretical model analysis, this study empirically tests determinants of the yield of the risky assets in GEM.Empirical result shows that CAPM in the Growth Enterprise Market is not effective, which is different from some recent domestic results. According to the test result, there is no direct relation between beta and stock return. Empirical analysis of Fama-French Three-Factor Model indicates that the three factor model has high fitting degree and applicability in GEM. However, the explanation power of those pricing factors is different, Market beta is the strongest, HML follows, and SMB is the weakest. Meanwhile we find that small firm effect and value effect do exist in the sample period. In other words, small-caps stocks have performed better than large-cap historically and value stocks have performed better than growth stocks historically. Empirical test on HAM shows that HAM is more available than the prior two models and the effects of Market Risk Premium, Adjust Cost, Book to Market Ratio and Enterprise’s Performance on stock returns are significant.更多还原