The relative-relative liquidity premium and cross-sectional returns: Evidence from China

We propose an aggregate liquidity factor using a principal component analysis approach to measure different aspects of illiquidity information in Chinese markets. We aggregate information on 7 liquidity-related trading-volume- and price-volatilitybased characteristics to build the new illiquidity factor (LiqAR). To eliminate the sentiment and moment information introduced into the LiqAR during the aggregating process, we standardized the LiqAR in both horizontal and vertical directions. We evaluate various factor pricing models incorporating the relative-relative LiqAR factors and find that the new LiqAR factor generates significantly higher cross-sectional stock returns than convenience illiquidity factors. The results remain robust after introducing different pricing factors, such as size, book-to-market ratio, and earnings-to-price ratio. The new 2, 3, and 4 factor pricing models can account for numerous return anomalies.本文采用主成分分析(Principal Component Analysis)方法构建总流动性因子，用以测度中国市场中非流动性信息的多维度特征。我们整合7项与流动性相关的、基于交易量和价格波动率的特征指标，搭建全新的非流动性因子LiqAR。为消除聚合过程中引入LiqAR的情绪与动量信息，我们从横向与纵向两个维度对该因子进行标准化处理。本文对纳入相对-相对LiqAR因子的多类因子定价模型开展检验，结果显示：新的LiqAR因子相较于常规非流动性因子，能够产生显著更高的横截面股票收益。在加入市值、市账率、盈价比等各类定价因子后，上述结论依然保持稳健。本文所构建的2因子、3因子及4因子定价模型，可解释诸多收益异象。