Cross-section Of Stock Returns Anomalies And Investor Sentiment

Cross-section stock returns anomalies are persistent and systematic patterns that efficient market hypothesis can not explain.Nowadays,there are so many papers about anomalies however most of these papers only focus on one single cross-section stock return anomaly,but fail to treat these anomalies as a whole.In the paper,I survey 58 popular variables which comes from 6 categories such as momentum,value-versus-growth,investment,profitability,analyst and trading friction.By using common methods of long-short portfolio formation,I analyze the correlations among these variables and find many anomalies.Further I find that anomalies are mostly from the momentum and trading friction categories.The results are still robust when I control the Fama-French three factor sand five factors.In addition,equal-weighted gets us more anomalies than value-weighted which corresponds to Hou(2016).Behavior finance assume investors have irrational belief and preferences.Therefore,the noise trader matters due to the existence of limit of arbitrage.And investor sentiment is the core of behavior finance.In this paper,we construct a sentiment composite to be the proxy of investor sentiment.We find that investor sentiment which is proxies by a sentiment composite index is positive correlated with some anomalies abnormal returns.This means that anomalies perform greater in high sentiment period than in low sentiment period.Further,we find the phenomenon exists because that the long part of the anomaly performs better in high sentiment period.