DOC《摘要》

摘要 看跌风险度量具有直觉上的优越性,它更关注资产收益向下方向的运动.

为了刻画投资者在股市牛熊阶段的不同风险偏好,本文引入了转换风险度量的风险度量方式. 在一定假设条件下,均值下偏矩框架的组合选择问题得出与经典均值方差分析一样的理论结果,在市场均衡时得到类似于MV CAPM的MLPM CAPM. 为了将重要的经济意义条件加诸于定价模型,本文的实证模型选用定价核框架,并运用Hansen-Jagannathan的方法构造了统计量来比较不同风险度量下的CAPM.由于不同的市场替代可能导致不同的实证结论,文章用三种加权方法计算市场收益,而为了保证收益数据的一致性,分组时采用与市场替代相对应的加权方法构造组合. 估计结果对三种模型的排序印证了文章的设想.转换风险度量定价模型的表现符合理论语气,能较好的刻画投资者的风险偏好,而看跌风险度量因放弃太多信息而没有显示出直观的优越性.为了查看估计结果的稳健性,文章还用不同的无风险利率观察利率敏感性,并在适应性估计方法下比较三个模型,结果都进一步确认了转换度量的优越性. 关键词:资本资产定价;

看跌风险;

转换度量;

贝塔;

下偏矩 ABSTRACT Downside risk measures have an intuitive superiority, since it focuses on the downside movement of asset returns. In order to capture the investors' different risk preference in different market conditions, this paper introduces a switching risk mea-sure. Given some assumptions, I can obtain MLPM CAPM in the mean lower partial moment framework, as MV CAPM in the mean variance analysis. For the purpose of imposing economic conditions, I phrase my empirical research in the pricing kernel framework, and I construct a statistics to compare the CAPMs in the spirit of Hansen-Jagannathan approach. Since the different market proxy may lead to different result, this paper use three different weighting schemes. To guarantee the consistency of the returns data, I use the same weighting scheme as the market proxy when constructing benchmark portfolios. The rankings of the three different CAPMs con?rm the paper's idea. Even though the downside risk measure does not exhibit its intuitive superiority due to discarding too much information, the performance of switching measure meets the expectation. It captures the risk preference of the investors better then the other two risk measures. I then use two different risk free interest rates to check the sensitivity, and compare the the three CAPMs using ?tted estimating approach. The results con?rm the advantage of switching risk measure. KeyWords : CAPM;

Downside Risk;

Switching Measure;

Beta;

LPM