Academic journal article Journal of Applied Finance

Overnight Return, the Invisible Hand Behind Intraday Returns?

Academic journal article Journal of Applied Finance

Overnight Return, the Invisible Hand Behind Intraday Returns?

Article excerpt

An efficient market (weak form) will contain no significant price pattern, a view supported by numerous empirical studies. Our study, however, reveals a very strong negative autocorrelation between overnight and intraday returns, regardless of our sampling method or the methodology in use. Though this poses potential market mispricing opportunities, we conclude that future studies are needed in order to determine whether anyone other than a market maker can fully exploit these opportunities.

(ProQuest: ... denotes formulae omitted.)

In Fama's (1970) survey, almost none of the studies offer strong counterevidence to the "efficient markets" hypothesis for the stock market. Since then, however, studies of actual market performance reveal a number of apparent inconsistencies, including the January effect (Bhardwaj and Brooks, 1992; Thaler, 1987a; Thaler, 1987b), day-of-theweek effect (French, 1980; Kamara, 1997), other seasonal effects (Ariel, 1987; Cadsby and Ratner, 1992; Lakonishok and Smidt, 1988; Rozeff and Kinney, 1976; SEC Litigation Release 19033, 2005), size effect (Reinganum, 1981), value line enigma (Stickel, 1985), low price-to-earnings (PE) effect (Banz, 1981; Basu, 1977; Reinganum, 1981), and a variety of other anomalous security market behaviors such as value verses growth stock performance (La Porta, Lakonishok, Shliefer, and Vishny 1997; Agrawal and Tandon, 1994).

Our study of the daily returns for stocks listed on the NYSE, AMEX, and NASDAQ reveals a potentially anomalistic relationship that may have significant market implications. According to the weak form of the efficient market hypothesis, which is supported by numerous studies, a time series of returns should either be uncorrelated or the magnitude of any possible correlation should be too small to be financially meaningful. Our finding, that overnight stock returns are strongly autocorrelated with subsequent intraday returns, presents evidence that is inconsistent with the weak form of the efficient market hypothesis.

The current study was inspired by Branch, Ma, and Sawyer (2010), a study on how closed-end funds' returns were related to their net asset value (NAV) performance. Branch et al. (2010) bifurcate the daily returns of closedend funds into overnight returns and intraday returns and find that the two exhibit a strong negative autocorrelation. The present study also expands the study by Branch and Ma (2008), which uses panel data analysis to uncover a strong autocorrelation between overnight returns and intraday returns for stocks listed on three major stock exchanges.

Our study confirms and extends in a more detailed way the finding of negative autocorrelation in stock returns as exhibited in Branch and Ma (2008) and Branch et al. (2010). We model our analysis in the mode of the Fama-MacBeth (1973) two-stage method. Our results are statistically significant and robust.

The remainder of this paper is organized as follows. We review the literature in Section I. Section II discusses the data sample and methodology. Section III reports results. Section IV discusses possible causes for autocorrelation in stock returns, including a brief discussion of literature on specialist behavior. Section V discusses practical implications of our findings, and Section VI concludes.

I. Literature Review

A number of issues related to the relationship between overnight and intraday returns are considered in the market microstructure literature. For example, Hong and Wang (2000, p. 299) cite the following: "empirical patterns.. .a, intraday mean return and volatility are U-shaped; b, intraday trading volume is U-shaped; c, open-to-open returns are more volatile than close-to-close returns; d, weekend returns are lower than weekday returns; and e, returns over trading periods are more volatile than returns over non-trading periods."

Stoll and Whaley (1990) report that open-to-open returns are more volatile than are close-to-close returns. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.