Annals of the University of North Carolina Wilmington International Masters of Business Administration.

Transcription

1 Annals of the University of North Carolina Wilmington International Masters of Business Administration

2 THE DAY-OF-THE-WEEK TRADING EFFECT: A COMPARISON OF THE CROATIAN AND SLOVENIAN MARKET Hekuran Murati A Thesis Submitted to the University of North Carolina Wilmington in Partial Fulfillment of the Requirements for the Degree of Master of Business Administration Cameron School of Business University of North Carolina Wilmington 2010 Approved by Advisory Committee Nivine Richie Luther Lawson Clay Moffett Chair Accepted by Dean, Graduate School

3 TABLE OF CONTENTS ABSTRACT... iii ACKNOWLEDGMENTS... iv LIST OF TABLES... v LIST OF FIGURES... vi INTRODUCTION... 1 The Croatian Market... 3 The Slovenian Market... 4 LITERATURE REVIEW... 6 DATA DESCRIPTION... 9 METHODOLOGY... 9 Section A. Entire Period Regression Analysis Section B. Three Phases Regression Analysis Section C. Currency Change Comparison RESULTS Section A.- Entire Period Regression Analysis Section B.- Three Phases Regression Analysis Section C. Currency Difference Comparison SUMMARY AND CONCLUSIONS REFERENCES APPENDIX ii

4 ABSTRACT Existing studies on the day-of-the-week anomaly focus mainly on the US market and other developed markets, but little attention has been given to developing markets. This paper focuses on an empirical analysis of the day-of-the-week effect anomaly using returns from two south-east European stock market indices, Croatia and Slovenia. The research results show evidence of negative Monday returns but only Croatia is statistically significant. Days with positive return that are statistically significant are Thursday for Croatia and Friday for Slovenia. The research also shows that the day-of-the-week effect behaves differently across market phases. In addition, research on the Slovenian market indicates a fading day-of-the-week effect after the adoption of the Euro as a currency. iii

5 ACKNOWLEDGMENTS My thanks go to Irwin Metzger, my finance professor from Rochester Institute of Technology, who introduced me to the field of finance during my undergraduate studies. I am especially grateful to Jens Hermsdorf, finance professor at the Hochschule Bremen University of Applied Sciences, who helped me refine my knowledge about finance studies. Many thanks go to Karen Barnhill and Barbara Hoppe, members of the Graduate Programs Office in the Cameron School of Business of the University of North Carolina Wilmington, for their excellent assistance during my work on the thesis. Special thanks go to my parents who supported me along the way, morally and financially. I hope I ve proven worthy of their effort and patience. Finally, I would like to thank my committee members, Dr. Clay Moffett, Dr. Nivine Richie, and Dr. Luther Lawson, for their continuous guidance and assistance throughout the process. iv

6 LIST OF TABLES Table Page 1. Summary Statistics on Daily Returns of the two Markets Time Intervals and the Number of Observations for each Market Phase Croatia: Entire Period Slovenia: Entire Period Croatia: Consolidation Phase Slovenia: Consolidation Phase Croatia: Bullish Phase Slovenia: Bullish Phase Croatia: Bearish Phase Slovenia: Bearish Phase Slovenia: Pre-Euro Slovenia: After-Euro...25 v

7 LIST OF FIGURES Figure Page 1. Croatia s Index CROBEX and the 100-day Moving Average in red Color Slovenia s Index LJSEX and the 100-day Moving Average in red Color...13 vi

8 INTRODUCTION A topic that is largely debated in the finance world is the Efficient Market Hypothesis (EMH). This theory states that it is unlikely for investors to beat the market on a consistent and risk adjusted basis. However, evidence shows that there are market anomalies that support otherwise. One such anomaly is the phenomenon called day-of-the-week effect. Many empirical studies on this effect show that the average daily return of the market is not the same across all days of the week, contrary to the EMH. Based on empirical results, this effect is present not only in developed economies like the US, but also in emerging markets. The most important characteristic of the day-of-the-week effect, as a result of numerous studies in the field, seems to be the abnormally positive average returns on Fridays, and the abnormally negative average returns on Mondays, also known as the weekend effect. This effect seems to be present in most of the western markets including US, UK, and Canada. However, in some Asian markets like Japan, the highest negative returns tend to be on Tuesday. The general consensus regarding the cause of the negative Mondays is that during the weekend, a lot of negative news is released that affects the companies, thus, resulting in a lower opening price. Others try to explain this through the psychological lens, stating that Monday is the first weekday of work. Hence, investors are not very happy and sell more, which drives prices down. Whereas on Friday, they are happy and optimistic because it is the last working day of the week, and as a result there will be more buying relative to selling, which drives prices up. On the other hand, the Tuesday negative returns are mostly explained by the assumption that the weekend negative news that affect the US market on Monday, are reflected the next day on the countries that are in a time-zone ahead of that of the US. However, none of these

9 explanations are fully representative for the abovementioned effects, thus, leaving more room for additional research on this topic. The purpose of this study will be to examine whether the day-of-the-week effect is present in the Croatian and Slovenian markets and see how both markets compare with each other. A previous study done by Patev, Lyroudi, and Kanaryan (2003) in other countries in the Central European region including Romania, Latvia, Czech Republic, Slovakia, Slovenia, Poland, and Hungary shows that there is evidence of this phenomenon in some of these countries, with the exception of Poland and Slovakia. However, no such studies have been done so far for the Croatian market, which makes an interesting topic to study. In addition to that, this paper is motivated to see the comparison of the two neighboring countries, Croatia and Slovenia. In contrast to Croatia, Slovenia is a member of the EU since May 1, 2004, and has adopted the Euro as a currency since January 1, This consecutively means that it has easier and broader access to capital from other EU members. In order to add more value and provide a more detailed insight into the behavior of the market return in different environments, the data set will be divided into 3 intervals, each representing a phase of both markets: the Consolidation Phase when the market has just come out of a Bearish period ad has no strong direction; the Bullish Phase where the market crosses the 100-day moving average from below and turns strongly bullish, and the Bearish Phase where the market crosses the 100-day moving average from above and turns strongly bearish. These phases will then be studied separately to determine how the day-of-the-week effect behaves through all three periods. A similar study done by Arora and Das (2007) in the Indian market shows that, although the results for the overall interval were positive in regards to the day-of-the-week effect, the 2

10 study of the three phases separately gave different results. Only the Consolidation phase provided evidence of the day-of-the-week, whereas the Bullish and Bearish phase did not. In addition to the two sections presented above, this paper will also examine how the dayof-the-week effect compares between a pre-euro Slovenia and an after-euro Slovenia. The paper will try to analyze whether there is a difference between the two periods, having into account that with the adoption of the Euro the Slovene Stock Exchange was made available to Euro zone investors as well. This due to the fact that the currency exchange risk is non-existent anymore for potential investors. Following will be a summary of the economic state of both countries, Croatia and Slovenia, as well as details about their Stock Exchanges. The Croatian Market Croatia is a country located in the southeastern part of Europe, also known as the Balkans. Croatia borders Slovenia and Hungary to the north, Bosnia and Herzegovina to the south and Serbia to the east. It has access to the Adriatic Sea, which separates it from Italy, and also enables a good climate for tourism development. Tourism accounts for a significant portion of the Croatian economy. In 2008 they received 11 million tourists, which is more than double their population of 4.5 million. In the world rankings of the most popular destinations, Croatia takes the 18 th place. Other dominant industries are shipbuilding, food processing, and the chemical industry. According to IMF, for the year 2008, Croatia had a GDP of $69.3 billion, which translates to a GDP per capita of $15,633. 3

11 The Croatian stock exchange is located in Zagreb, the capital of the country, after which it is also named: Zagreb Stock Exchange (Croatian: Zagrebacka Burza). ZSE was established in 1991 and it trades shares of Croatian companies as well as bonds and commercial bills. As of November 9, 2010, the exchange has 376 listed shares, 56 listed bonds, and 13 listed commercial bills. It publishes two indices, CROBEX for stocks, and CROBIS for bonds. The exchange works 5 days a week, Monday through Friday and open hours are 10:00 am to 4:00 pm. The trades are done in the local currency, the Croatian Kuna. The Slovenian Market Slovenia, also referred to as Republic of Slovenia, is located in Central Europe, just north of Croatia. Other countries that Slovenia also borders are Italy to the west, Hungary to the east, and Austria to the north. It also has access to the Adriatic Sea in the southwest. Since 2004, Slovenia has been a member of the European Union (EU), and since 2007 it has adopted the EU zone currency, the Euro. Slovenia s main industries are construction and the service industry. The service industry dominates the country s output with about 57% of the share. A large portion of the service industry comprises of mainly financial services. Slovenia is also heavily dependent on trade. Its trade volume, including imports and exports is at 120% of GDP, which makes the Slovenian economy very dependent on its trading partners. Slovenia mainly trades with other countries of Central Europe, including Germany, Austria, Italy, and France. Slovenia s GDP for the year 2008, according to the IMF, was $59.4 billion, which is lower than its neighbor, Croatia. However, if we take into account the GDP per capita, Slovenia 4

12 comes on the high-income countries with $29,472 per capita for the year 2008, which is nearly double that of Croatia. The Slovenian Stock Exchange is located in the capital of the country, Ljubljana, and its original name is Ljubljanska Borza (LJSE). It was established on December 26, Since June 2008, the LJSE is owned by the Vienna Stock Exchange. The operating hours of the exchange are from 9:30 AM to 1:00 PM, with pre-market sessions from 8:00 AM until 9:30 AM. Trading days include all weekdays, excluding Saturdays, Sundays, and other holidays as announced by the LJSE in advance. The LJSE publishes 4 indices, which are the Slovene Stock Exchange Index (SBI 20, since March 2010 name was changed to LJSEX), Slovene Blue chip Index (SBI TOP), Investment Funds Index (PIX), and Bond Index (BIO). As of November 9, 2010, the LJSE has 179 listed securities, out of which, 80 are stocks, 11 are ETF and 88 are bonds. This paper will proceed with the following sections: Section 2 will be a review on the literature and prior research done on market anomalies, focusing on the day-of-the-week effect. Section 3 will develop the methodology used to calculate the results of the study as well as give insight on the data that will be used and their source. Section 4 will be discussion and analysis of the results from the model. And finally, section 5 will a conclusion of the paper summarizing the findings throughout the process. 5

13 LITERATURE REVIEW According to Maberly (1995), the day-of-the-week effect was first documented by Kelly (1930), to be re-discovered again by Cross (1973) who says that large stock market decreases tend to occur between the Friday close and the Monday close. Since then, French (1980), Lakonishok and Levi (1982), Keim and Stambaugh (1984), Jaffe and Westerfield (1985), Harris (1986) and many others have found significant evidence that the US stock market returns from the Friday close until opening of Monday trading on average are negative. In addition, Gibbons and Hess (1981) noted similar results for bonds. Based on many other studies conducted, the day-of-the-week effect is present in other markets as well. Kramer and Runde (1993) provide evidence of a day-of-the-week effect in the German market. Alexakis and Xanthakis (1995) give evidence of such an effect on the Australian, Canadian, and Japanese markets. Solnik and Bousqet (1990) find similar results in the French market. Ho (1990) in the stock markets of Hong Kong, Korea, Malaysia, New Zealand, Philippines, Singapore, Taiwan, and Thailand. Nath and Dalvi (2004) also find such an effect in India. Most of the cases experience day-of-the-week effect on Fridays and Mondays. However, some countries have Tuesday as the day with the lowest return. Such examples would be Kato (1990) for the Japanese market where he observes that Tuesday returns are the lowest. Other examples would be Aggarval and Rivoli (1989) in the markets of Hong Kong, Singapore, Malaysia and Philippines. Also the stock market in Turkey has average negative returns on Tuesdays according to Balaban (1995), and Bildik (1997). So mainly the days where abnormal returns are evident are usually Monday, Tuesday and Friday. However, other scholars have revealed that there is no day-of-the-week effect like in the case of Spain Gardezabal and Regulez

14 (2002), or Kato (1990) observes the highest return on Wednesdays for the Japanese market. On the other hand, there is the case of the Greek market where during both Tuesdays and Wednesdays had negative returns according to Nikou (1997), whereas during Mondays, Wednesdays and Fridays had positive returns on average according to Lyroudi, Subeniotis and Komisopoulos (2002). In some recent studies of the US market however, the day-of-the-week effect seems to be fading or even non-existent. This comes as a consequence of the market becoming efficient after it became aware of the anomaly. One such study that demonstrates a fading day-of-the-week effect for the US market is Schwert (2003). In his paper, he also points out that other effects such as the size effect and value effect have also disappeared from the US market. On the other hand, other studies show that stock exchanges in some emerging markets still show signs of the day-of-the-week effect, leading us to think that in the emerging markets that stock exchanges are not as efficient as the more mature ones like the US. As mentioned in the introduction, an explanation for the weekend effect is considered to be the tendency for bad news to be announced during the weekend, so when the market opens on Monday, it opens lower. However, according to Chen and Singal (2003), speculative short sales account for a partial explanation of the day-of-the-week effect that has high return Fridays and low return Mondays. The inability of the speculators to trade over the weekend, causes them to close their short positions on Friday, and re-establish new short positions on Monday, thus, causing stock prices to rise on Fridays and fall on Mondays. They support their hypothesis through evidence based on a comparison between high short-interest stocks and low short-interest stocks, stocks 7

15 with and without actively traded options, IPO s, zero short-interest stocks, and high volatility stocks. Further evidence that supports this theory is provided by Lakonishok and Maberly (1990) who observe a relative increase in transactions on Mondays. They find a tendency for individuals to increase the number of sell transactions compared to buy transactions, which can partly explain the weekend effect. Wang, Li and Erickson (1997) demonstrate that the weekend effect occurs primarily in the last two weeks, the fourth and the fifth week of the month. They also indicate that the weekend effect in the NASDAQ index, which is mainly comprised of smaller companies, is stronger than the weekend effect of the NYSE-AMEX, which on the other hand is dominated by large companies. Their findings are also supported by Gibbons and Hess (1981), and Keim and Stambaugh (1984). If this stands true also for emerging markets like Croatia, then we should expect a very strong day-of-the-week effect on Fridays and Mondays due to the fact that most companies enlisted with the Zagreb Stock Exchange are small-cap. As presented above, a lot of studies have been done on the day-of-the-week effect and its variations. Studies also show that this effect can also be present in emerging markets like Croatia. Therefore, it is of high interest to perform an examination of the Croatian stock market in order to gain more insight regarding its functioning and whether it performs similar to its neighboring countries which use the euro, unlike Croatia. 8

16 DATA DESCRIPTION The time interval for both countries for the extracted data to be studied is from January 17, 2000 through September 17, For the entire period extracted, there are 2,663 observations for Croatia and 2,662 observations for Slovenia. The difference in observations, although the interval is the same for both countries, comes as a consequence of the holidays in the two countries. The data for the Croatian Market was extracted directly from Zagreb Stock Exchange s website. The data obtained include the daily prices of the ZSE index CROBEX which is the official ZSE share index and has been published since September 1, 1997 and is weighted by free float adjusted market capitalization. Its base value was set on The data for the Slovenian Market was extracted from Ljubljana Stock Exchange s website. The data obtained include the daily prices of the LJSE index called LJSEX which measures the performance of the entire LJSE stock market. Its base value was set on METHODOLOGY The daily index returns for both countries are computed as follows: R it = log (I it / Iit-1 ) * 100 (1) Where: is the daily percentage return of stock index i on day t, and represent the closing price on day t and t-1 for the same index.

17 Table 1. Summary Statistics on Daily Returns of the two Markets Croatia Slovenia Mean Standard Error Standard Deviation Sample Variance Kurtosis Skewness Range Minimum Maximum Count Section A. Entire Period Regression Analysis In order to examine whether there is a day-of-the-week effect for the entire period, a regression analysis will be run for both data sets. This paper will use the same model employed by Gibbons and Hess (1981), Jaffe and Westerfield (1985), Kramer and Runde (1993), and Bayar and Kan (2002): R t = Intercept + a 1 TUE + a 2 WED + a 3 THU + a 4 FRI + e t (2) Where: R t is the trading day s index return, Intercept represents Monday when all other dummy variables are 0, TUE dummy variable takes 1 if the day is Tuesday, 0 otherwise, WED dummy variable takes 1 if the day is Wednesday, 0 otherwise, THU dummy variable takes 1 if the day is Thursday, 0 otherwise, FRI dummy variable takes 1 if the day is Friday, 0 otherwise, e t is the error term and represents Monday when all other dummy variables are 0. 10

18 The dummy variables presented in the model above are considered explanatory factors in order to examine two research questions: (i) whether there is a day-of-the-week effect; (ii) whether the effect is significant at a level of 5%. In order to improve the data, an outlier filter of -4> >4 was applied to the index returns of the Croatian Market. The filter removed three outliers, two positive Mondays and one negative Friday. Section B. Three Phases Regression Analysis By breaking the data set into 3 subsets, the intention is to see whether there is seasonality in the day-of-the-week effect for the Croatian market, as well as how the result compares to the original data set. As presented in the study done by Arora and Das (2007), the day-of-the-week effect can behave differently depending on what phase the market is. Therefore, this paper will also examine how this effect exhibits in the three periods separately by running a regression analysis using the same model as used for the entire data set above in Section A. The three periods for both stock markets were determined by using a 100-day moving average for the daily closing prices of two main indices, CROBEX and LJSEX. The 100-day moving average crosses three times the daily prices of the indices to determine the three phases: Consolidation; Bullish; and Bearish. For the Croatian Market these three phases start at January 17, 2000 and end at March 27, 2009, whereas for the Slovenian market these three phases start at January 17, 2000 and end at March 13, Following, on Table 1 are presented starting dates and the number of observations for all three phases. 11

19 Table 2. Time intervals and the number of observations for each market phase Croatia Slovenia Start Observations Start Observations Consolidation Phase 17-Jan Jan Bullish phase 30-Aug Apr Bearish Phase 3-Dec Nov Figure 1 and Figure 2 show graphs of both market s daily index prices represented with the blue color, as well as the 100-day moving average line represented with the red color. Figure 1. Croatia s index CROBEX and the 100-day moving average in red color 6, Croatia 5, , , , ,

20 Figure 2. Slovenia s index LJSEX and the 100-day moving average in red color Slovenia /14/2000 5/25/2000 9/28/2000 2/9/2001 6/20/ /25/2001 3/11/2002 7/19/ /26/2002 4/8/2003 8/18/ /22/2003 5/4/2004 9/7/2004 1/12/2005 5/23/2005 9/28/2005 2/6/2006 6/16/ /20/2006 3/7/2007 7/18/ /23/2007 4/8/2008 8/14/ /19/ /5 / /8 /2009 1/1/ /26/2010 Section C. Currency Change Comparison This part of the analysis will focus on the currency change that happens in Slovenia. With the Euro Adoption Act, Slovenia changed its currency from the Tolar to the Euro on January 1, 2007 with a two-week dual currency circulation to be fully adopted by January 15, This date will be the benchmark for creating another data sub-set. By applying a regression analysis to the data sub-set using the same model as in Section A of methodology, this paper intends to study whether a currency change affects the day-of-the-week effect, and whether the effect is significant. 13

21 RESULTS The regression results of the index prices of both countries for the period from January 17, 2000 until September 17, 2010 are shown in 3 following sections. Section A will present the results intending to explain the day-of-the-week effect for the overall period including both countries. Section B will present the regression results of the same data series, but split into 3 phases - consolidation, bullish, bearish as shown above in methodology, intending to compare the behavior of the day-of-the-week effect across these phases. Section C will present the regression results from the period when Slovenia adopted the Euro as a currency, intending to explain any changes in the day-of-the-week effect comparing the Euro and non-euro countries. Section A.- Entire Period Regression Analysis The regression results presented in Table 3 show that there is evidence of abnormal returns for all days of the week in the Croatian market. All days are statistically significant at the 5% level. The estimates of the returns for the entire period show Monday as having a negative return that is statistically significant, while the returns of the other days are all positive with Thursday being the highest. Table 4 presents the regressions results for the entire period on the Slovenian market. The estimates of the returns for the entire period show Friday as the only day having a statistically significant positive return at a 5% level. This partially confirms previous studies done on other emerging markets which show that the day-of-the-week effect has a presence. However, the difference is that in the Croatian market all days show a statistically significant day-of-the-week effect whereas in the Slovenian market only Friday. This is in line with other studies that claim that the day-of-the-week fades with the

22 increasing efficiency of the market. One reason for this difference could also be the adoption of the Euro in Slovenia. Section B.- Three Phases Regression Analysis Consolidation Phase The regression results for the estimates of the returns presented on Table 5 show that none of the days is statistically significant at the 5% level for the Consolidation Phase in the Croatian market. Table 6 shows only Friday is statistically significant at the 5% level for the Slovenian market. These findings do not confirm the study done by Arora and Das (2007) where they show that the Consolidation Phase provided evidence of the day-of-the-week effect for the Indian market. Bullish Phase Table 7 shows that there is evidence of abnormal positive returns for all days of the week for the Croatian market. However, none of the days are statistically significant at the 5% level. Table 8 shows that there is evidence of abnormal positive returns for all days of the week for the Slovenian market. However, only Thursday is statistically significant at the 5% level. These findings partially confirm the study by Arora and Das (2007), where they show that during the Bullish Phase the stock market does not provide any evidence of the day-of-the-week effect. Bearish Phase Table 9 shows that there is evidence of abnormal returns for all days of the week for the Croatian market. However, only Monday is statistically significant at the 5% level. Table 10 15

23 shows that there is evidence of abnormal returns for all days of the week for the Slovenian market. However, none of the days are statistically significant at the 5% level. These findings for the Bearish Phase partially confirm the study by Arora and Das (2007), where only Slovenia shows no evidence of statistically significant day-of-the-week effect. Section C. Currency Difference Comparison The data sub-set representing the Slovenian Market before the Euro adoption shows evidence of abnormal returns on all days of the week, as presented on Table 11. However, only Friday is statistically significant at the 5% level. On the other hand, the data sub-set representing the Slovenian Market after the Euro adoption shows evidence of abnormal returns on all days of the week, as presented on Table 12. However, this time none of the days is statistically significant at the 5% level. This confirms the expectations that the Euro adoption by the Slovenian Government had an effect on the stock market. The assumption is that by adopting the Euro, Slovenia opened up to investors from the Eurozone as well since there was no more currency-exchange risk and as such, contributed to a fading day-of-the-week effect with the increased market efficiency. 16

24 SUMMARY AND CONCLUSIONS Various studies have been performed on market anomalies, one of the being the day-ofthe-week effect. However, despite quite extensive analysis on the developed markets, there influence of the day-of-the-week effect on smaller emerging markets is not sufficiently explored. This paper analyzes the day-of-the-week effect on two small emerging markets, Croatia and Slovenia, by using empirical research data from the Zagreb Stock Exchange and Ljubljana Stock Exchange from the time interval 17 January 2000 to 17 September The task of the paper was to define whether the day-of-the-week effect stands valid for both these countries by running a regression analysis on the daily returns of the Indices for the overall period. The research revealed that there was evidence of the day-of-the-week effect on both markets. The effect was strongest on the Croatian market with all days of the week being significant, whereas Slovenia had only Friday statistically significant. These results are consistent with previous studies done on emerging markets. The second task was to analyze if the day-of-the-week effect behaves differently across different phases of the markets. The phases being: Consolidation; Bullish; and Bearish. To do this, the initial data set was divided into three smaller sub-sets using a 100-day moving average to determine a switch to the next phase. The research revealed a partial confirmation of Arora and Das (2007) where Croatia shows a significant day-of-the-week effect only on the Bearish phase, whereas Slovenia shows a significant day-of-the-week effect on the Consolidation and Bullish phases. The third research task deals with the question whether the adoption of the Euro as a currency had an influence on the day-of-the-week effect on the Slovenian market. The analysis is done by dividing the initial set of data for the Slovenian market into two sub-sets, the first being

25 data before the Euro adoption, and the second being the data after the Euro adoption. The regression analysis revealed that there was evidence of the day-of-the-week effect for the first sub-set, but not for the second. This shows a tendency of a fading day-of-the-week effect on the Slovenian market after the Euro adoption. The regression analyses were generally consistent with the most prominent previous studies. According to the research results, there is evidence that the day-of-the-week effect has a tendency to fade in smaller markets as well, as long as they continue to improve efficiency. The tendency could be supported by the increasing globalization of stock markets like the case of Slovenia. 18

26 REFERENCES 1. Aggarwal and Rivoli (1989). On the Relationship Between the United States' and Four Asian Equity Markets. Asean Economic Bulletin, No. 6, Alexakis and Xanthakis (1995). Day- of- the- Week Effect on the Greek Stock Market. Applied Financial Economics, No. 5, Arora and Das (2007). Day of the Week Effects in NSE Stock Returns: An Empirical Study. Working Paper 4. Balaban (1995). Day - of- the- Week Effects: New Evidence From an Emerging Stock Market Applied Economics Letters, No. 2, Bayar & Kan (2002). Day of the Week Effects: Recent Evidence from nineteen Stock Markets. Central Bank Review, Vol. 2, No. 2 (July), Bildik (1997). "Hisse Senedi Piyasalarinda Aykiriliklar: IMKB'nda Gun Etkisi", Unpublished Working Paper at Finance Department of Istanbul University 7. Chen & Singal (2003). Role of Speculative Short Sales in Price Formation: The Case of the Weekend Effect. The Journal of Finance, Vol. 58, No. 2, (Apr.), Cross (1973). The Behavior of Stock Prices on Fridays and Mondays. Financial Analysts Journal, Vol. 29, Issue 6 (Nov./Dec.), French (1980). Stock Returns And The Weekend Effect. Journal of Financial Economics, Vol. 8, March, Gardeazabal and Regulez (2002). The Weekend-Dividend Effect in the Spanish Market, Presentation at the European Finance Management Association, Annual Conference, June 2002 in London, UK 11. Gibbons & Hess (1981). Day of the Week Effects and Asset Returns. Journal of Business, Vol. 54, No. 4 (Oct.), Ho (1990). Stock Return Seasonalities in Asia Pacific Markets. Journal of International Financial Management and Accounting, Vol. 2, No. 1 (Spring), Harris (1986). A Transaction Data Study of Weekly and Intradaily Patterns in Stock Returns. The Journal of Financial Economics, No. 16, Jaffe & Westerfield (1985). The Week-End Effect in Common Stock Returns: The International Evidence. Journal of Finance, Vol. 40, No. 2 (June),

27 15. Kato (1990). Weekly Patterns in Japanese Stock Returns. Management Science, Vol. 36, No. 9 (Sep.), Keim & Stambaugh (1984). A Further Investigation of the Weekend Effect in Stock Returns. Journal of Finance, Vol. 39, No. 3 (July), Kelly (1930). Why You Win Or Lose, Fraser Publishing Krämer & Runde (1993). Kalendereffekte auf Kapitalmärkten Eine empirische Untersuchung für deutsche Aktien und den DAX. Zeitschrift für betriebswirtschaftliche Forschung, Special Issue 31, Lakonishok & Levi (1982). Weekend Effects on Stock Returns: A Note. Journal of Finance, Vol. 37, No. 3 (June), Lakonishok & Maberly (1990). The Weekend Effect: Trading Patterns of individual and institutional Investors. Journal of Finance, Vol. 45, No. 1 (Mar.), Lyroudi, Subeniotis, and Komisopoulos (2002). " Market Anomalies in the A.S.E.: The Day of the Week Effect", Presentation at the 2002 European Finance Management Association, Annual Conference, June 2002 in London, UK 22. Maberly (1995). Eureka! Eureka! Discovery of the Monday Effect belongs to the ancient Scribes. Financial Analysts Journal, Vol. 51, Issue 3 (Sept./Oct.), Nath & Dalvi (2004). Day of the Week Effect and Market Efficiency Evidence from Indian Equity Market Using High Frequency Data of National Stock Exchange. Paper 24. Nikou (1997). Market Results and the Phenomenon of the Weekend Effect on the Stock Market Returns, Master Thesis, University of Macedonia 25. Patev, Lyroudi, and Kanaryan (2003). The Day of the Week Effect in the Central European Transition Stock Markets. Tsenov Academy of Economics Finance and Credit, Working Paper No Prokop (2008). Once Upon a Time: The Day-of-the-Week Effect in German and US Stock Market Returns. 21st Australasian Finance and Banking Conference, Working Paper 27. Schwert (2003). Anomalies and Market Efficiency. Handbook of the Economics of Finance, Edited by G.M. Constantinides, M. Harris and R. Stulz 28. Solnik and Bousquet (1990). Day - of- the- Week Effect on the Paris Bourse. Journal of Banking and Finance, Vol. 14,

28 29. Wang, Li & Erickson (1997). A new Look at the Monday Effect. Journal of Finance, Vol. 52, Issue 5 (Dec.),

29 APPENDIX Appendix A. Tables with regression results from the study. Table 3. Croatia: Entire Period Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations 2660 Table 4. Slovenia: Entire Period Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations 2662

30 Table 5. Croatia: Consolidation Phase Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations 1153 Table 6. Slovenia: Consolidation Phase Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations 1551 Table 7. Croatia: Bullish Phase Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations

31 Table 8. Slovenia: Bullish Phase Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations 397 Table 9. Croatia: Bearish Phase Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations 325 Table 10. Slovenia: Bearish Phase Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations

32 Table 11. Slovenia: Pre-Euro Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations 1743 Table 12. Slovenia: After-Euro Coefficients Standard Error t Stat P-value Intercept Tuesday Wednesday Thursday Friday R Square Adjusted R Square Significance F Observations