总结了几十种知产定价模型因子的定义

Anomaly CharacteristicsAnomaly definitions and descriptions are heavily based on the lists of characteristics compiledby Hou et al. (2015); Kogan and Tian (2015); McLean and Pontiff (2016); Novy-Marx andVelikov (2016). All accounting variables are properly lagged. For annual rebelancing, returnsfrom July of year t to June of year t+1 are matched to variables in December of t−1. Returnsfrom January to June of year t are matched to variables in December of year t−2. Financialvariables with a subscript “Dec” below are computed using the same timing convention.For monthly rebalancing, returns are matched to the latest quarterly report lagged onemonth. Additional lagging (if required) is reported for each variable below individually.All subindices below are measured in months. Time subscript t refers to time at which aportfolio is formed.1. Size (size). Follows Fama and French (1993). size = MEJun. The CRSP end of Juneprice times shares outstanding. Rebalanced annually.