Long Term Dependence of Popular and Neglected Stocks

In this study, we establish a connection between the levels of market attentions of a stock with its long memory features. We construct two portfolios of US equities based on Doyle et al’s (2006) criteria for neglected and popular stocks and measure the degrees of persistence for their daily returns from January 1, 2003 to December 31, 2007. We find that all stocks except for one display anti-persistence in the neglect portfolio; while the popular portfolio stocks uniformly display random walk returns. This suggests that there is a connection between the persistence features of stock return series and the levels of “neglect” of stocks. We use book to market ratio, analyst coverage, and transaction frictions to classify the levels of market neglect of stocks. Based on our study, while these criteria combined appear to contribute to the long memory features of daily returns of stocks, we also suspect the presence of other factors driving the persistence of stock returns