The role of individual investors' gambling preference in January effect.

摘要

Rozeff and Kinney (1976) were the first to End that stocks in NYSE provide abnormally large returns in January. Following researchers such as Keim (1983), Branch and Chang (1990), and Bhardwaj and Brooks (1992), examine this "January effect" and find that small firms and low price stocks are more prone to this effect. The January effect is so robust that it is not a unique phenomenon in U.S., but also occurs in other countries (e.g., Gultekin and Gultekin, 1983; Kato and Schallheim, 1985; Tong, 1992); In addition, it not only occurs in the stock market, but also in the bond market (e.g., Chang and Pinegar, 1986; Maxwell, 1998; Starks, Yong, and Zheng, 2006) and in option market (Doran, Jiang, and Peterson, 2012).;The current literature provides two main hypotheses to explain the disproportionate January returns in financial markets: the tax-loss selling hypothesis (Wachtel, 1942) and the window dressing hypothesis (Haugen and Lakonishok, 1987. Both theories address the illiquidity factor of individual and institutional investors at the end of the year. In this dissertation, I revisit the January effect in the U.S. stock market, and examine the role of investors' appetite for lottery-type stocks in asset pricing in January.;I utilize five variables -- stock price (PRC), idiosyncratic skewness (ISKEW), idiosyncratic volatility (IVOL), maximum daily return (MAX), and the George and Hwang ratio (GH-Ratio) -- as measures of stocks' lottery features.;Partitioning stocks into quintiles sorted on these lottery-type characteristics, I find that lottery-type stocks outperform in January from the period of 1965 to 2008, and this outperformance is more significant among past loser stocks. In order to control other variables that may explain stock returns, I also employ Fama-MacBeth (1973) regressions to re-examine the findings, and the results are consistent. Next, to address the issue of individual investors' gambling preference, I equally divide the sample into three sub-samples based on stocks institutional ownership. I find abnormally greater performance from lottery-type stocks among past losers in stocks with the lowest institutional holdings (Group IHL), but not in stocks with the highest institutional holdings (Group IHH). Moreover, the greater performance of lottery-type stocks persists up to six months beyond January. Lastly, I investigate the implication of the "other January effect" (Cooper, McConnell, and Ovtchinnikov (2006)) on lottery-type stocks. The results indicate that investors are more confident to invest in lottery-type stocks in years with positive January market return.