在第一篇文章裡，我們主要檢定二個解釋小公司及價值溢酬的因子模型。在CAPM的架構下，Ferguson and Shockley (2003)認為小公司(small size)及價值型股票(high bookto-market ratio)的溢酬是因為選用了股票市場的投資組合來當作市場投資組合的替代所造成的，真實的市場投資組合應包含市場上所有的股票及債權。Ferguson and Shockley (2003)提出兩個與公司的相對槓桿及相對危機相關的因子，發現這兩個因子能涵蓋Fama and French(1993)三因子模型的解釋能力。然而，Ferguson and Shockley (2003)的實證結果可能受到變數衡量誤差及投資組合形成的影響。本文採用Brennan et al. (1998)的方法來檢驗Fergusonand Shockley (2003)的因子是否能完全解釋市場上的異常現象，包含小公司及價值溢酬。Brennan et al. (1998)的方法是採用個別股票來分析，因此不受變數衡量誤差及投資組合形成問題的影響。我們的實證結果發現不論是Ferguson and Shockley (2003)的模型或是Fama and French (1993)的三因子模型，都無法完全解釋價值溢酬。我們進一步發現，由Ferguson and Shockley (2003)及Fama and French (1993)所構成的五因子模型，可以完全解釋價值溢酬。 第二篇文章將個別股票的流動性風險應用到資產定價的一些重要議題上。過去文獻發現流動性較差的股票具有較高的期望報酬，因此提出流動性溢酬的異常現象。本篇文章採用文獻上使用最為廣泛的三種流動性指標：Datar et al. (1998)用來衡量交易量面向的指標、Amihud(2002)用來衡量價格影響的指標以及Liu (2006)用來衡量交易速度的指標，對造成流動性溢酬的可能因素進行廣泛的探討，檢驗其背後的因素是來自理性的因子模型或偏向行為面的特徵模型；我們採用Daniel and Titman (1997)的方法，發現整體而言因子模型是較能解釋流動性溢酬的。 This study contains two essays on factors, characteristics and asset-pricing anomalies. Essay 1: Do Relative Leverage and Relative Distress Really Explain Size and Book-to-Market Anomalies? In a CAPM framework, Ferguson and Shockley (2003) argue that size and value premiums are due to the improper use of an equity-only market portfolio. To complement the debt claims in the market portfolio, they propose two factors on relative leverage and relative distress, and show that the two factors subsume the explanatory power of Fama and French’s (1993) SMB and HML factors. However, based on an errors-in-variables free methodology proposed by Brennan, Chordia and Subrahmanyam (1998), we find that neither Ferguson and Shockley’s (2003) nor Fama and French’s (1993) models fully explain the book-to-market anomaly. As a further application, we find an augmented five-factor model which incorporates Fama and French’s (1993) factors into Ferguson and Shockley’s (2003) factors, is able to capture the book-to-market anomaly. Essay 2: What Drives the Liquidity Premium: Factors or Characteristics? In this paper, we investigate whether the nature of the liquidity premium is driven by the risk-based model or the characteristics-based model. We provide a comprehensive analysis by using three widely adopted liquidity measures, including the trading quantity dimension of Datar et al. (1998), the price impact dimension of Amihud (2002) and Liu’s (2006) measure which emphasizes on the trading speed dimension. By applying the methodology of Daniel and Titman (1997) and Davis et al. (2000), we show that overall the liquidity premium is better explained by the risk-based factor model.
