Calendar Effects on Security Price Anomalies - Literature review Example

?Topic: Calendar effects (Seasonal ties). January Effect, the Day of the Week Effect, the Holiday Effect Introduction Calendar effects on security price anomalies have been a hot issue of interest among academicians, economists, statisticians and market experts for many decades. Not all but some of these anomalies are widely referred as calendar effects. Calendar anomalies such as the day of the week effect, monthly or January effect, the trading month effect, and holiday effect in stock exchange markets have perplexed financial economists for more than 50 years. Literature Review For the day of the week effect in stock market returns, Gibbons & Hess (1981), Fama (1991), (Grossman and Stiglitz 1980)), French (1980)), Lakonishok and Levi (1982)), Rogalski (1984)) and Keim and Stambaugh (1984)), Harris (2002), Lakonishok and Smidt (1988), Allen and Karjalainen (1993) have exhibited the impression of this phenomenon. However, according to Kenourgios & Samitas (2008) latest global level studies indicate that this market irregularity is loosing its sheen to the extent of showing no symptoms of visibility or its effect has considerably reduced in developed stock markets since it was first registered in the 1980s (e.g., Chang et al., 1993; Schwert, 2001; Steeley, 2001; Kohers et al., 2004; Hui, 2005)). Generally, it is taken for granted that the apportioning of stock returns is the same for all days of a week but it is not that crucial an assumption for achieving market equilibrium. Empirical research indicates unequal stock returns on different days of the week, particularly weekend effects on Monday returns. It could be because Monday’s return is computed over three in stead of one day causing the mean and variance to be higher than other days, which could be approximately three times higher. Although Fama (1965)) has compared daily mean returns but he has found Monday’s difference on stock returns to be 20% more than other daily returns, which is similar to what other researchers have focussed on (Gibbons & Hess, 1981). Background At the ground level, the theory of efficient capital markets is the same i.e. the theory of competitive equilibrium used on asset markets. It is based on the Ricardian principle of comparative advantage. The same notion in financial markets is applied with the only assumption of getting competitive advantage because of the gaps in information, not “fully reflected” in prices, thus, setting the basis for profitable trading rules (LeRoy, 1989). Fama (1991) has reviewed market efficiency literature by selecting the relevant research in the previous 20 years. He has taken the market efficiency hypothesis, which simply means that security prices are totally predictable as based on given information. It works on the assumption that information and the trading costs are mostly 0 (Grossman and Stiglitz 1980)). Another assumption of the efficiency hypothesis is that prices indicate information to the level where marginal benefits of accruing profits on the basis of information are less than the marginal cost (Jensen 1978)). There is so much opaqueness that one cannot measure market efficiency due to various versions of market efficiency to be reflected by going back on any type of efficiency hypothesis within the given information and the trading costs. There are other models that present a straightforward approach towards this issue. French (1980) also tested two models on stock returns as based on calendar time hypothesis and trading time hypothesis for a period between 1953 and 1977 for daily returns on Standards & Poor’s composite portfolio not matching with the models, indicating positive average return for the four days while negative return for Monday. Harris (2002) has studied weekly and intra-day designs in stock returns by employing transaction data for large and smaller firms. Findings indicate negative Monday close-to-close returns from the Friday close to Monday open for larger firms while for small firms it happens majorly the same for the whole of Monday trading. For all firms irrespective of firm size, differences in intraday returns are big in the initial 45 minutes after the market starts functioning. Prices come down on Monday morning and go up on other weekday mornings. Generally, the design of intraday returns is the same on all weekdays except on Friday when there is an increase in stock prices. Lakonishok and Smidt (1988) have performed the tumultuous task of researching 90 years of daily data of Dow Jones Industrial Average to record the proof of seasonal designs in the rates of returns, which they have found indicating regular disturbed returns at the change of the week, at the change to the new month, at the change of the year and near to holidays. By using a uniform data base and methodology, they have found proof of such seasonal irregularities in stock returns. For calculation they have studied the daily closing prices of Dow Jones Industrial Average from 1897 to 1986, reflecting on the behaviour of the U.S. security market index. They did perform a test on monthly, semi-monthly, weekend, holiday, end-of-December, and turn-of-the-month seasonality. Specialty of their research has been longer periods of data testing than conducted in previous research, giving a touch of completeness to their research. Criticality Aspect of Literature Research Getting meaningful results through repeated data has been one of the major drawbacks of research made by different academics although data sources such as Centre for Research in Security Prices (CRSP) and COMPUSTAT have ended the insufficiency in empirical research on stock prices. Therefore, the danger of data snooping, one of the generic considerations, (boredom and noise being other generic considerations), in the conducted academic research is significant. It stresses on the fact that the quality of the proof could be compromising. Merton (1985)) has pointed towards the possibility of giving unwanted significance to research that report such irregularities. Such studies overestimate the proof of the presence of irregularities in the behaviour of stock market. Rose (1986)) has also studied this partiality in publishing only such studies that project more than actual findings (Lakonishok and Smidt, 1988). Lakonishok and Smidt (1988) have observed anomalies nearer to the turn of the week, the turn of the month, the turn of the year and when holidays are approaching, particularly, the rate of return on Monday is considerably negative (-0.14 percent), the price hike near the change of the month crosses the total monthly price growth, the price growth from the last trading day before Christmas to the finishing of the year goes over 1.5 percent and the rate of return before holidays exceeds to beyond 20 times the normal rate of return. The theory hypothesis that these specific time-period anomalies could have taken place accidently cannot be ruled out, but this is quite rare a possibility. No regular monthly design was observed in the returns or any regular behaviour for returns in the initial part of the month to be higher. Considering the extent of anomalies, Lakonishok and Smidt (1988) have related it to the size of the tick, which is very small change in price of 12.5 cents. As average price on the NYSE for a share is near to $40, a shift of one tick affects a change of 0.313% or more, which is quite bigger than most seasonal irregularities. For instance, the average Monday price down of-0.144% comes well within one tick. Even if we ignore the extent of these irregularities, their recurrent happening demands an answer to the issue and attention gets diverted to the processes that fix prices in securities market. Having a single explanation for all the anomalies is not a possibility. Academics have pointed towards the role of inventory settings of various traders such as [Rock (1989) and Ritter (1988)], the timing of trades by knowledgeable and non- knowledgeable businesses [Admati and Pfleiderer (1988a)], and experts' strategies against knowledgeable traders [Admati and Pfleiderer (1988b)]. Other reasons that could be attributed to seasonal irregularities are related to the timing of corporate news releases [Penman (1987)], seasonal designs in cash flows of peoples and institutional investors, tax-inspired trading [Lakonishok and Smidt (1986)] and the window dressing inspired by regular monitoring of portfolio managers [Haugen and Lakonishok (1988) and Ritter and Chopra (1989)]. Allen and Karjalainen (1993) have used genetic algorithms to find whether there exists a possibility of the application of technical trading rules on Standards and Poor’s Composite Stock Index in 1963-89, affecting the returns in comparison to ordinary buy-and-hold strategy. They have compared their results with the standard models of a random walk, an auto-regressive model and a GARCH-AR model. They have examined the results attained through genetic algorithms by conducting bootstrapping simulations and traditional empirical tests. Their study analysis indicates that returns have been greater than expected both figure-wise and economically too although transaction costs have been included in deriving returns. Allen and Karjalainen (1993) find no dependable explanation of the reduced returns at times when trading rules are robust in the market. There can be other risk parameters but it can not be summarily rejected a notion that abnormal returns could be due to greater possibility of risks. If more updated risk factors are found, genetic algorithms can be leveraged by investors to include those factors into their trading strategies. But we should not forget that additional returns accruing from technical rules are not sudden as the break-even transaction costs come under the costs borne by big institutional investors. Allen and Karjalainen (1993) have not paid attention to details such as timing of intraday trading rule signals and the occurrence of big returns. In sum, technical rules cannot assist smaller investors to get the leverage from such tactics as it helps large institutional investors in equity markets. In a way, the outcomes derived by Allen and Karjalainen (1993) match with Grossman and Stiglitz’s (1980) improved view of market efficiency that making profits via information increase returns which cover costs in the long period. Kenourgios & Samitas (2008) research on the day of the week effect on return for the leading index of Athens Stock Exchange (ASE) offers a different perspective as it reflects on the non-US Stock exchange market paradigms. They have included the past work in their conditional differential framework done on the Greek stock market. They have found the effect of the day of the week in the ASE during 1995-2000. Interestingly, they found the effect reducing after Greece becoming a member of the euro zone. It is the result of competitive change and institutional remedies adopted by ASE. Special about the research done by Kenourgios & Samitas (2008) is the add-on to the current literature by using both mean and variance guidelines for a developing stock market of ASE by covering period of research further from 1996-1997 that included crucial happenings in the political, macro economic and stock market environment of Greece wile enriching the experience of equity markets entering the European Union off late. Cao (2006) has studied the existence of the day of the week effect, the monthly effect and the holiday effect on index returns by using the Shenzhen Stock Exchange A Share Index and Shanghai Stock Exchange A Share Index covering a 11 years period starting from 1995 to 2005. Ass per the study outcomes, the day of the week effect, the monthly effect and the holiday effect (only in Shanghai) exist as calculated through the GARCH (1, 1) model, the GARCH-M model and the Modified GARCH (1, 1) model. The results regarding the days of the week interestingly are not the same as they are in American, Greek or European financial markets, as highest returns are visible on Tuesday and Friday and the lowest return rates are seen on Monday and Thursday. In Chinese stock markets, monthly extraordinary outcomes happen in the months of March and January and reduced returns appear in September and December in Shenzhen Stock Exchange and Shanghai Stock Exchange but the holiday effect is only visible in the Shanghai Stock Exchange. Conclusion A review of literature starting from a time period covering Gibbons & Hess (1981), Fama (1991), Lakonishok and Smidt (1988), Allen and Karjalainen (1993)till Harris (2002) and finally reaching Kenourgios & Samitas (2008) while including Cao (2006) along makes this literature review global in the true spirit by considering Dow Jones Industrial Average, Athens Stock Exchange, Shenzhen Stock Exchange and Shanghai Stock Exchange. All the above academics have affirmed the presence of calendar effect in the financial markets by reviewing the traditionally acquired data although data sets of DJIA cover a span of 90 years but Lakonishok and Smidt (1988) agree to the research drawback of reusing the obsolete data and pointed towards the risks of data snooping, affecting the quality of evidence, which in itself could be a serious manifestation on the genuineness of outcomes. We at the same time should not ignore the prevalent trend of ignoring the opposite viewpoint of giving undue significance to the concept, as stated by Merton (1985)). Certain drawbacks of the technical trading rules, as stated by Allen and Karjalainen (1993) need further investigation and critical review for not paying attention to details such as timing of intraday trading rule signals and the occurrence of big returns. The current literature review stresses on the fact that improved market efficiency helps in making profits via information as stated by Grossman and Stiglitz’s (1980). Calendar effects exist, there cannot be any doubt over it but there is need to further study the reasons behind their existence although there are clear signs that with the advancement in market efficiencies scope of security price anomalies would reduce in future as is evident from the Athens Stock Exchange getting updated with the entry in Euro Zone. At the same, we have seen that the Chinese stock markets have different days of recording calendar effects unlike their counterparts in other parts of the world as stated by Cao (2006), which stresses on the occurrence of the calendar effect, prompting further research on why it happens. References ALLEN, F., KARJALAINEN, R., 1993. Using genetic algorithms to find technical trading rules. Rodney L. White Center for Financial Research. The Wharton School, University of Pennsylvania. Available from: http://www.cenet.org.cn/userfiles/2010-3-22/20100322022117952.pdf [Accessed 5 April 2011]. CAO, Q., 2006. The Calendar effect on A-share index return in Chinese stock market. The University of Nottingham. Available from: http://edissertations.nottingham.ac.uk/547/1/06MAlixqc1.pdf [Accessed 5 April 2011]. FAMA, EUGENE F., 1991. Efficient capital markets: II. The Journal of Finance [online], 46 (5), 1575-1617. Available from: http://www.jstor.org/stable/2328565 [Accessed 5 April 2011]. FRENCH, K. R., 1980. Stock returns and the weekend effect. Journal of Financial Economics [Abstract], 8 (1), 55-69. Available from: http://calendar- effects.behaviouralfinance.net/weekend-effect/ [Accessed 5 April 2011]. GIBBONS, M. R., HESS, P., 1981. Day of the week effects and asset returns. The Journal of Business [online], 54 (4), 579-596. Available from: http://www.jstor.org/stable/2352725 [Accessed 5 April 2011]. HARRIS, L., 2002. A transaction data study of weekly and intra-daily patterns in stock returns. Journal of Financial Economics [online], 16 (1), 99-117. Available from: http://www.sciencedirect.com [Accessed 5 April 2011]. KENOURGIOS, D., SAMITAS, A., 2008. The day of the week effect patterns on stock market return and volatility: evidence for the Athens Stock Exchange. International Research Journal of Finance and Economics, 15. Available from: http://www.eurojournals.com/irjfe%2015%20dimitris.pdf [Accessed 5 April 2011]. LAKONISHOK , J., SMIDT, S., 1988. Are seasonal anomalies real? A ninety-year perspective. The Review of Financial Studies, 1( 4), 403-425. Available from: http://rfs.oxfordjournals.org/content/1/4/403.short [Accessed 5 April 2011]. LEROY, S.F., 1989. Efficient capital markets and martingales. Journal of Economic Literature, 27 (4), 1583-1621. Available from: http://www.jstor.org [Accessed 5 April 2011]. Read More