Print Reference This Reddit This

Study The Effect Of Holidays On Stock Return Finance Essay

Jacobs and N. Levy (1988) calendar anomalies have long been part of market tradition. Studies of the day-of-the-week, holiday and January effects first began to appear in the 1930s and although academics have only recently begun seriously to examine these return patterns, they have found them to survive close study. Calendar regularities generally occur at cusps in time the turn of the year, the month, the week, the day and the Holidays. They often have significant economic impact. For example, the "Blue Mon- day" effect was so strong during the Great Depression that the entire market crash took place over weekends, from Saturday's close to Monday's close. The stock market actually rose on average every other day of the week. (Jacobs and N. Levy, 1988)

Lakonishok and Smidt (1988) find that stock market returns for the days previous to holidays are significantly higher than returns for other days. (Lakonishok and Smidt, 1988)

Pettengill (1989) shows that while Preholiday and other day’s returns differ significantly, preholiday returns differ by holiday, firm size, and day of the week on which the holiday falls. Various theories have been proposed to explain the type of return generating process nearby holidays. Since holidays are often considered another type of market closing, similar to weekends, it seems reasonable that an explanation for one might help explain the other. For example, if the weekend effect could be explained by conclusion actions, then this case is relevant because holidays can delay conclusion for up to two days and could have an effect on returns up to a week previous to the holiday. (Pettengill, 1989)

Jacobs and N. Levy (1988) holiday differences appear fairly constant over time. In the most recent decade, however, pre-holiday returns have not been exceptional. But the effect does not appear to be a statistical object. The settlement process discuss as a potential explanation for the day of the week effect has complex implication for fluctuations in value around holidays. For example this theory predicts a high Thursday return proceeding a Friday holiday, which is what occurs. But it forecast a lower than average Friday return preceding a Monday holiday, and this is not consistent with practical results. Moreover, the size of any value changes occurring because of settlement procedures is much too small to account for the holiday effect. Abnormal pre-holiday returns are not attributable to increased risk. Another perspective is afforded by holidays not associated with market closings, like St. Patrick's Day or Rosh Hashanah. Such days do not experience abnormal returns. The absence of abnormal returns may be due to the lack of a trading break or to a lower level of event than that associated with major market holidays. (Jacobs and N. Levy, 1988)

Kim and J.Park (1994) one of the puzzling empirical findings reported in recent studies is the presence of abnormally high stock returns on the day before holidays. (Kim and J.Park, 1994)

Objective:

To find the Effect of Calendar holidays on the stock return.

Variables:

Variables are those stocks rates before the holidays and stock return in terms of capital gain.

Hypothesis:

H1: Stock returns on the day before holiday is higher than the day after holiday

Literature Review

The aim of the study to observe that what the Effect of Calendar Holiday is’s on Stock return. Stock return is higher before holiday or after the holiday. This Study is observing on Karachi Stock Exchange (KSE).

Former research on seasonality has found a relationship between the seasonal patterns of stock returns and firm size. Keim (1983) shows that the January effect is mainly a small firm effect. Also, Keim and Stambaugh (1984) find that the weekend effect is greater for small-firm stocks than for large-firm stocks. Based on the relationship of the above mentioned effects with firm size there are some possible explanations for these seasonal patterns have been proposed in the finance literature. In relation to the holiday effect, Keim (1989) suggests that systematic patterns in investor buying and selling behavior explain the unusually high returns observed on the trading days previous to holidays. If closing prices two days before the holiday tend to be recorded at the bid while closing prices on the trading day before the holiday are recorded at the ask, then these logical patterns would produce high returns observed on the trading day previous to the holiday. If this is true and then the trading pattern would be more important for small firm stocks since the relative bid-ask spread is larger for these stocks (Kim and Park, 1994)

Kim and Park (1994) recent studies also examine the relationship between holiday effect and firm size. Pettingill (1989) reports that small firms outperform large ones both on January and non-January preholidays. Ariel (1990), on the contrary, finds that there are no incremental preholiday returns accruing to small firms after adjusting day-of-the-week effect and excluding New Year's Day. To resolve this difference of findings, this paper investigates whether the holiday effect persevere across size decide set. (Kim and Park, 1994)

Kim and Park (1994) study of the U.S. stock market shows that the holiday effect exists uniformly across all three major markets and size deciles portfolios. A natural question that arises is whether the holiday effect is present in the stock markets of countries that have different holidays and institutional measures. If the holiday effect exists across countries in unkindness of the differences in holidays and institutional measures, the results would drop more light on possible causes of the holiday effect. (Kim and Park, 1994)

Kim and Park (1994) study holiday effect is observed in the U.S. S&P 500 index returns for the 1972-1987 periods. The holiday effect is also indication in the Japanese market for the same period. The mean return on the trading day before Japanese holidays is 0.1897 percent whereas the mean return during ordinary days is 0.0435 percent. The results of both the t- test and non parametric median test exhibit that the difference of the mean and median returns between preholidays and ordinary days is statistically significant. For the U.K. FT 30 index, the average return on the trading day before the U.K. holidays is 0.2228 percent, which is five times as much as the ordinary daily mean return of 0.0397 percent. However, the f-statistic for the difference of the mean returns between preholidays and ordinary days is slightly significant (p-value = 0.104), whereas the nonparametric test statistic exhibits a significant difference of the median returns between preholidays and ordinary days. The low r-statistic for the U.K. holiday effect could be due to larger standard deviation of the U.K. stock market returns and fewer observations of the U.K. holidays relative to the U.S. and Japan. (Kim and Park, 1994)

There are many empirical studies (e.g., Eun and Shim (1989), Hamao, Masulis, and Ng (1990), and Becker, Finnerty, and Gupta (1990)) that provide indication of international linkages of daily stock market returns. Also the international evidence of the day-of-the-week effect indicates an association between the stocks returns patterns of the U.S. and other markets (see Jaffe and Westerfield (1985a), (1985b)). Related to the international linkages of stock markets the holiday effects in the U.K. and Japanese markets are related with the U.S. holiday effect. Mainly this paper attempts to answer the following questions: 1) Are the holiday effects in the U.K. and Japanese markets related with the U.S. holiday effect And 2) Does the U.S. holiday effect bring high returns on nonpreholidays of the U.K. and Japan? (Kim and Park, 1994)

A. Ariel (1990) Despite the much higher return, the pre-holiday variance of return is no larger than the return variance for all other days; means and variances do not increase proportionately, as would be the case if the pre-holiday mean return, which is 9 to 14 times the mean for the other days, resulted from 9 to 14 "regular" days somehow compounded into one day. Rather, it seems an extra component of return is added to a regular trading day. Indeed, not only is the pre-holiday variance no greater than the variance for other days, the pre-holiday variance is actually lower than the variance of non- pre-holidays. This fact serves to emphasize that the high pre-holiday return is not a reward for bearing extra risk. (A. Ariel, 1990)

A. Ariel (1990) significant portion of the total twenty-year cumulative return earned by the market index can be recognized to the returns earned on pre-holidays. The mean pre-holiday return exceeds the mean return for all non-pre- holidays by factors of 9 and 14 for the equally and value weighted indices. Therefore the eight pre-holidays collectively equal 72 or 112 non-pre- holidays in their impact on annual returns. Since there are 251 trading days in the average year, holiday returns will constitute an insignificant fraction of the total return accruing to the indices. High returns predominate only on the single trading day previous holidays and not on other days around the holiday period. (A. Ariel, 1990)

Roll (1983a) observes that the period of January high returns in fact starts on the last trading day of December. This day is the trading day prior to the New Year and is thus one of the pre-holidays included in the above tests. Hence, the high January effect return on the pre-holiday before New Year's may be partially responsible for the statistical significance of the tests reported in Table I. As a check on this possibility, Table IV(A) reports a regression model in which a pre- holiday dummy variable is regressed against daily index returns, and Table IV(B) repeats the regression after adding a separate pre-New Year dummy variable in addition to the pre-holiday dummy. As shown in Table IV(B), the incremental return earned on pre-New Year's in excess of that earned on other pre-holidays is negligible for the value-weighted index but is large and statistically significant for the equally-weighted index, consistent with Roll's claim that small firm January effect high returns start on the day before New Year's.(Roll, 1983a)

J. Fabozzi, K. Ma, E. Briley (1994) pre holiday return is the difference between the close to close prices for the calendar days previous to the holiday. The post holiday return is a two day return which is the difference between the closing price of the calendar day previous to the holiday and that of the day immediately following the holiday. The non holiday return is the return over a one day period, without days surrounding holidays. The distinction between trading and non trading returns provides impending as to whether the higher return is associated only with trading in the market, or is a result of valuation at any rate of trading. The fact that the non trading return is also higher than usual may suggest the need of higher non trading returns to persuade traders to hold inventories over non trading periods. Since the higher pre holiday return goes beyond the impact of demand and supply of market trading, it implies that there is a valuation component in the higher return that does not depend on whether the trading takes place previous to holidays. Holidays when exchanges are closed may be considered another form of market closing, like weekends. Thus, higher returns previous to holiday closings are related to higher returns before the weekend and before the end of the day. To examine the holiday effect in the absence of the holiday closings, we compare the pre holiday returns for holidays when exchanges are open with the non holiday returns. (J. Fabozzi, K. Ma, E. Briley, 1994)

J. Fabozzi, K. Ma, E. Briley (1994) unlike other calendar time seasonalties, a holiday effect, if there is one, should be area definite. That is, a holiday is relevant only to the country or region where it is celebrated. Assets traded on the world markets around the clock most probably should be less affected by holidays unique to the United States, for example. The numerous international futures contracts in our sample allow us to test the significance of the holiday effect when assets are traded beyond borders or time zones. Addition of the trading hours should alleviate some of the unique trading pattern attributable to the information definite to a certain area. The distinction between international futures contracts and domestic futures contracts gives further insight into the merit of various hypotheses. While not prevent other possibilities, for international futures contracts where U.S. trading is not exclusive the portion of the higher pre holiday return resulting from inventory adjustment should be small, because the inventory risk over a non trading period is less for international futures contracts. At the same time, the portion of positive post holiday returns related with the favorable psychology should be conserved even in the case of internationally traded futures contracts. In the futures market, there is a considerably higher return for the day previous to a holiday. Since the positive holiday effect is associated more with the domestic exchange closed holidays, this would imply that the return effect may be related to the inventory adjustment process associated with market closings. It appears that investors are more reluctant to take positions, especially short positions instantly before holidays. (J. Fabozzi, K. Ma, E. Briley, 1994)

Apart from these two main seasonal effects, there are others like the December end

Holiday effect (Ariel, 1990) which finds that the pre-holiday returns are usually large.

Then there is the turn of the month effect (Ariel, 1987) where returns are found to be

higher around the end of the month. The Friday the Thirteenth effect suggests that returns on these days is negative as compared to other Fridays (Kolb and Rodriguez, 1987, Dyl and Maberly, 1988). Lamb et al. (2004) find that negative returns for the spring and fall daylight savings time weekend returns found by Kamstra et al. (2000) are not reliable and significant.

Armand Picou (1999) holiday effects are not new to the published literature. The holiday studies (Agrawal and Tandon, 1994; Ariel, 1990; Brockman and Michayluk, 1997; Kim and Park, 1994; Kim, 1988; Pettengill, 1989) all find increased returns during the day before a holiday period in many major markets. A response after the holiday is not found.

While the holiday effect is well recognized, what has not been recently examined is the current relationship of one exchanges holiday to other exchanges. The present study examines the dealings of all exchanges upon the return to trading and includes six major exchanges (Japan, Australia, Hong Kong, London, Canada and the USA). The large number of inter relationships subjective by differences in time zones attached with the use of daylight and overnight returns are especially insightful. (Armand Picou, 1999)

Van Der Sar (2003) using daily data on a value-weighted index of all shares in the Netherlands (1981 to 1998) found abnormally high returns in the second half of December and around the month and negative returns and higher instability on Monday. Alagidede and Panagiotidis (2006) found an April effect for Ghana stock prices contrary to the usual January effect. Pandey (2002) confirmed the existence of seasonality in stock returns in India and the January effect and that the capital market in India was wasteful, and hence, investors can time their capital investment in Indian stock market to improve returns. However, the magnitude of the January effect depends upon the country and the composition of the index (Hawanini and Keim as reported by Marquering, 2002). Though the end of year effect is more distinct for small firms’ stocks and as a result for equally weighted indices, it is also present in value-weighted indices (Marquering 2002).

Aswath Damodaran (1989) is consistent evidence that returns on Mondays are more negative than returns on any other day of the week. French (1980), examining daily stock returns between 1953 and 1977, notes that the mean return on Monday was negative and lower than the mean for any other day of the week during 20 of the 25 years. Keim and Stambaugh (1984) find negative Monday returns for the S&P 500 as early as 1928 and for firms of all sizes. Rogalski (1984) offers proof that much of the Monday effect between 1974 and 1983 was due to the price change between the close on Friday and the open on Monday and was so actually a weekend effect. Smirlock and Starks (1986) using hourly values of the Dow Jones Industrial Average over the 1963-1983 period wind up that the weekend effect has shifted from illustrate active trading on Mondays to illustrate the non trading weekend. Harris (1986) finds proof of a size effect in weekend returns using transaction data, with much of the negative returns accruing between Friday close and Monday open for large firms and between Friday close and Monday close for smaller firms. While the weekend effect may be too small to give rise to profitable trading policy, it cannot be easily clarified away. French compares Monday returns with returns after weekdays which are market holidays and wind up that the Monday returns are caused by some weekend effect rather than by a general closed market effect. Keim and Stambaugh find negative Monday returns using bid prices on actively traded OTC stocks and, therefore, reject expert related explanations. One hypothesis for the weekend effect is that firms release bad news toward or after the close on Friday because they fear panic selling on financial markets. For this to cause the negative Monday returns, there has to be a parallel market incompetence since rational investors should, over time, anticipate the bad news. (Damodaran, 1989)

Alagidede (2008) However, a growing number of studies suggest that betas of common stocks do not sufficiently explain cross sectional differences in stock returns. Instead, a number of variables, such as firm size, ratio of book to market, and price/earning ratios, that have no source in extant theoretical models, seem to have significant predictive ability. For example, Basu (1977) and Banz (1981) found that the ratio of price to earnings and market capitalizations of common equity, respectively, provided significantly more descriptive power than beta. Also, stock returns are found to be systematically higher or lower depending on the time of the day, day of the week, month of the year and Holidays. (Alagidede, 2008)

Alagidede (2008) the month-of-the-year and turn-of-the-month effect postulates that returns are estimated to be higher in the month of January, and especially, in the first few trading days of the month than other months of the year. Over the years, evidence show that returns observed on days previous a public holiday are on average many times greater than returns on other trading days.(Alagidede, 2008)

Alagidede (2008) These regularities in stock returns, otherwise known as calendar anomalies (effects), have occupied experimental research on asset pricing models for nearly half a century, and present an inconsistency in experiential finance: Their realities shed hesitation on the validity of asset pricing models and hence challenge the belief in stock market efficiency. For case, investors could buy stocks on days (months) with unusually low returns and sell on days (months) with unusually high returns. Further, if the pre-holiday effect holds, it is possible to work out strategies that would yield returns over and above buy and hold. These would be inconsistent with the efficient markets hypothesis. But as their discovery seasonal patterns in stock returns have failed to yield consistent returns over and above buy and hold strategies. (Alagidede, 2008)

As French (1980), Board and Sutcliffe (1988), Draper and Pauydyal (1997), Brooks and Persand (2001), Mills and Coutts (1995) dispute any trading rules resulting from the outlook of anomalies (effect) will be more than offset by the round trip transaction costs and illiquidity. Thus small calendar specific anomalies need not abuse any arbitrage circumstances. Further, it has been argued that even if there are no calendar specific effects, an extensive search (mining) over a large number of possible seasonalities is likely to yield something that appears to be an ‘anomaly’ by clean chance (see Lo and MacKinlay, 1990, Sullivan et al, 1999 and Burton, 2003).

Alagidede (2008) for nearly half a century of their discovery in markets world wide, there has been little evidence regarding African markets. The novelty of the paper rests on the following: (a) we test for the existence of two calendar anomalies in African indices— month-of-the-year and pre-holiday effects. African markets have a variety of institutional features that differentiate them from one another and from the markets in industrial and other emerging economies. The search for seasonality or other anomalies in the returns of African markets can provide important information on the role of institutional features on return behavior. This information may help stock exchange and regulatory authorities when they make policy decisions; (b) the paper explicitly accounts conditional heteroscedasticity in the month-of-the-year effects. (c) the question of whether trading rules can yield profits over buy and hold by exploiting seasonal patterns is explored. (Alagidede, 2008)

The definition of holidays varies among researchers (Brockman and Michayluk, 1998).

One definition looks at days, other than Saturday or Sunday, upon which the market is closed (Lakonishok and Smidt, 1988). Alagidede (2008) on the other hand, this prohibit exceptional events such as the end of apartheid in South Africa, the recent widespread political crises in Kenya that caused the market to close to traders, and natural disasters like hurricanes, etc which can cause unexpected shutting of markets. Moreover, some holidays e.g., Easter and most religious holidays which follow the lunar calendar change over time. To this end, we define the holiday effect as the return from the pre-holiday close to the post-holiday close. In other words, the holiday returns are the daily returns for the trading weekday that follows a non-trading weekday. Calendar effects are now accepted stylized facts in stock markets world-wide. However, the research on African stock markets regarding this issue is virtually non-existent. The pre-holiday effect is only significant for South Africa. There are high and significant returns in days previous a holiday. (Alagidede, 2008)

Mustafa (2005) Calendar anomalies are one of the features of financial market which is against the capable market hypothesis. Many researchers investigate the calendars anomalies which are based on Gregorian calendar. On the other hand different countries and societies also follow their own calendar which is based on religion in addition to Gregorian calendar. For example, Jewish society follow Hebrew calendar, which strictly based on luni-solar, the Christen society follows Gregorian calendar, which based on solar, Hindu and Chinese follow their own calendar. Muslim society follows the Islamic calendar, which is based on a lunar calendar, referred to as the Hijri calendar. This calendar contains twelve months that start with the appearance of new moon. The average days in a lunar month contain only 29.53 days that is why Islamic year is approximately eleven days shorter than the Gregorian year. In these religious calendars, there are religious days and month, which these societies observed e.g. Christen society celebrated X-miss days, Deepvali by Hindu society, and Vesk day by Buddhist. Like these religious days Muslim societies also observed and celebrated the religious month like as Ramdhan and Aashoora and day like Eid-ul-Fitar and Eid-ul-Azha. After celebrated Eid-ul-Fitar, the prices become normal. In month of Zil-Hajj, people slaughtered the animal like as cow, goat, camels’ etc, to follow the Sunnat-e-Ibrahim. It increases the consumption of people, which reduces the purchasing power hence saving decreases. Aashoora come after Eid-ul Azha, which observed as mourning month. (Khalid Mustafa, 2005)

Mustafa (2005) though a lot of consideration has been given on calendar anomalies but a little attention is given on the religious calendar effect on stock markets. For example, religious holidays effect on S&P500 index and NYSE trading volumes Frieder and Subrahmanyam (2004). These holidays paying attention on the Jewish High Holy Days of Rosh Hashanah and Yom Kippur and the Christian holy day of St. Patrick’s. They reported that volume declined on Rosh Hashanah and Yom Kippur, and that prices tended to increase during the two days that precede Rosh Hashanah and St. Patrick’s. (Khalid Mustafa, 2005)

Mustafa (2005) has been given little attention to the Islamic calendar effect. However, some studies are existing on the impact of Ramadan on stock returns. Alper and Aruoba (2001) analyze different macroeconomic variables in Turkey and show that the usual seasonal adjustment procedures based on fixed holidays often fail to remove all seasonality when the cycle are subject to moving holidays like Ramadan. They did not find any significant Ramadan effect in Istanbul stock market. Moreover, Husain (1998) analyzes the Ramadan effect in Pakistani stock market and verified that instability is significantly lower during the weeks of Ramadan. He does not find any significant changes in average returns during Ramadan. However, he did not compare the mean average return before and after Ramadan. Seyyed, Abraham, and Al-Hajji (2005) investigated the Ramadan effect in Saudi Arabian stock market. They analyzed several segment indices in the market and showed that instability and trading movement gone significantly during Ramadan. Their findings are matching to Husain (1998) findings i.e. they did not find any significant change in average returns during Ramadan and did not look before and after Ramdhan. (Khalid Mustafa, 2005)

Need help with your literature review?

Our qualified researchers are here to help. Click on the button below to find out more:

Literature Review Service

Related Content

In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: