Day-of-the-week effects: another evidence from top 50 Australian stocks

Griffith Research Online https://research-repository.griffith.edu.au Day-of-the-week effects: another evidence from top 50 Australian stocks Author Liu, Benjamin, Li, Bin Published 2010 Journal Title European Journal of Economics, Finance and Administrative Sciences Copyright Statement Copyright 2010 EuroJournals Publishing, Inc. The attached file is reproduced here in accordance with the copyright policy of the publisher. Please refer to the journal's website for access to the definitive, published version. Downloaded from http://hdl.handle.net/10072/35064 Link to published version http://www.eurojournals.com/ejefas_24_08.pdf

European Journal of Economics, Finance and Administrative Sciences ISSN 1450-2275 Issue 24 (2010) EuroJournals, Inc. 2010 http://www.eurojournals.com Day-of-the-Week Effects: Another Evidence from Top 50 Australian Stocks Benjamin Liu Griffith Business School, Griffith University, Brisbane, QLD 4111, Australia E-mail: b.liu@griffith.edu.au Tel: +61-7- 3735 3549 Bin Li Griffith Business School, Griffith University, Brisbane, QLD 4111, Australia E-mail: b.li@griffith.edu.au Tel: +61-7- 3735 7117 Abstract In this paper, we study day-of-the-week effects in the top 50 Australian companies across different industry sectors. Unlike other Australian studies, we study weekday seasonality using stock return data of individual companies. Utilizing the daily data for the period of January 2001 through June 2010, we find that weekday anomalies are mixed across companies and industries. We also find the largest mean weekday returns occur on Monday for 15 companies, most of which are the materials and energy companies. Further tests indicate that returns on Monday are significant larger than the other four days for six companies. Our results lend some support to the view of reversing weekend effects (e.g., Connolly, 1989; Doyle and Chen, 2007). Keywords: Australian stock market, market efficiency, market anomaly, day-of-the-week effect JEL Classification Codes: G14 1. Introduction Seasonal or calendar anomalies in equity markets (over specific days, weeks, months, and even years) have attracted a widespread attention and considerable interests among practitioners and academics alike. Over the last hundred years, a vast number of the literature from both the practitioner and academic fields studies day-of-the-week effects or day seasonality on returns of various assets, such as stocks, debt securities, futures, foreign currencies and even commodities. The earliest research can be traced back to as early as the late 1920s (Pettengill, 2003). There are more than two hundred published papers on this topic from different perspectives by 2010. Calendar anomalies, relying on the assumption that a certain pattern of stock markets is formed on the basis of the past stock price, can be used to predict the future stock price. If the pattern is fixed, informed investors can utilize the pattern to earn a risk-free profit by trading the stocks. The study of seasonality implies that investors could employ the findings on anomalies to predict the future behavior of prices (Fama, 1965). Certainly, seasonal anomalies are in contradiction to any form of efficient market hypothesis (EMH), particularly the weak-form efficiency.

79 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) However, during the last three decades, many researchers have documented evidence on seasonality of stock markets around the world. For example, Cross (1973), French (1980), Gibbons and Hess (1981), Keim and Stambaugh (1984), and Linton and Whang (2007) all report significantly negative average returns on the US stock markets on Mondays. Similar anomalies are found in international equity markets as well. The study includes Hindmarch, Jentsch, and Drew (1984), Jaffe and Westerfield (1985), Chang, Pinegar, and Ravichandran (1993), Tong (2000), Cai, Li and Qi (2006), and Lim, Ho and Dollery (2010). In contrast to the above findings, some recent literature, however, reports the declining or reversing weekend effects and a different shift of the weekend effect on large-capital securities and small capital securities (e.g., Connolly, 1989; Chang, Pinegar, and Ravichandran, 1993; Kamara, 1997; Doyle and Chen, 2007; and Worthington, 2010). Although Australia has a high level of share ownership, both institutional and individual, compared to other developed countries, market anomalies in the equity market are obviously underresearched. There are a few Australian studies on stock market seasonality, such as Keef and McGuinness (2001), Easton and Faff (1994), Marrett and Worthington (2008), Davidson and Faff (1999) and Worthington (2010). Their findings are mixed, and the results are sensitive to the sample period and the portfolios used. However, all the studies are only limited to the use of portfolio data and none of them use individual stock data. To address this issue, in this study, we investigate Australian equity seasonality using the top 50 companies stocks from for the period of June 2001 through June 2010. The rest of the paper is organized as follows: Section 2 review the relevant literature on day-ofthe-week effect. Section 3 offers a description of the data and its summary statistics. Section 4 describes empirical approaches and discusses empirical findings. Section 5 concludes this paper. 2. Literature Review The earliest research on return seasonality can be traced back to as early as the late 1920s (Pettengill, 2003). Up to 2010, there are more than two hundred published papers on this topic from different perspectives. In this section, we review the relevant literature on this topic. Early research is mainly concentrated on the US stock markets (see the detailed discussions in Pettengil, 2003). For example, Cross (1973), French (1980), Gibbons and Hess (1981), Keim and Stambaugh (1984) all report significantly negative average returns on the stock markets on Mondays. French (1980) studies the S&P 500 Index over the period 1953 through 1977, and Gibbons and Hess (1981) study the S&P 500 Index and CRSP value- and equally-weighted indexes for NYSE and AMEX securities over the period 1962 through 1978. Keim and Stambaugh (1984) extend the sample to actively traded OTC securities in addition to the S&P 500 Index. Linn and Lockwood (1988) investigate a large sample of OTC securities. Furthermore, Bessembinder and Hertzel (1993) study an even earlier sample period, covering the period from 1885 to 1989. Siegel (1998) examines the Monday effect over the period 1885 through 1997. Cho, Linton and Whang (2007) use a stochastic dominance approach and find strong evidence of the Monday effect for the period of January 1, 1970 to December 31, 2004. However, they find that the effect has reversed or weakened in the Dow Jones and S&P 500 indexes after 1987, but is still strong in more broadly based indexes such as the NASDAQ, the Russell 2000 and the CRSP indexes. They document a pervasive finding of abnormally high average Friday returns and significantly negative Monday returns on the US market. In contrast to the above findings, some recent literature reports declining or reversing weekend effects and a different shift of the weekend effect on large capitalization securities and small capitalization securities. Connolly (1989), who studies the weekday effect over the period 1963 through 1983, finds a significant difference between the Monday returns and the other weekday returns in the period prior to 1974, but no significant difference after 1974 although the average Monday return remains negative, which is confirmed by Chang, Pinegar, and Ravichandran (1993). Kamara (1997) finds that the Monday returns of large-firm stocks become positive but there is no similar

80 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) change for small-firm stocks. Mehdian and Perry (2001) find that for the pre-1987 period average Monday returns are negative and significantly lower than the other weekday returns for all five indexes, but for the post-1987 they find that average Monday returns are positive for large firms. Sullivan (2004) and Gu (2004) report similar findings, and Doyle and Chen (2007) find that the pattern of day seasonality on the stock markets are not fixed but varying over time. There is also some evidence of day seasonality in other equity markets. Jaffe and Westerfield (1985) find low Monday returns in the Canadian, British, Japanese, and Australian stock markets. Condoyanni, O Hanlon, and Ward (1987) find significantly negative Monday or Tuesday returns for seven developed markets (13 European countries, US and Canada, Australia, Hong Kong, Japan, Malaysia, New Zealand, Singapore, Mexico and South Africa). Chang, Pinegar, and Ravichandran (1993) find significantly negative Monday returns in 13 of 23 international markets, although their results are sensitive to the choice of statistical testing procedures. Dubois and Louvet (1996) provide further evidence of the existence of low Monday returns for developed markets (Canada, the US, Japan, Hong Kong, Australia, Germany, France, the UK and Switzerland) in an examination of eleven indexes from nine countries during the period 1969 through 1992. They find that returns are lower at the beginning of the week (negative Tuesday returns for the Australian (1980-1992) and Japanese markets (1969-1988)). They further conclude that the anomaly disappears for the most recent period in the US but the effect is still strong for European countries, Hong Kong and Canada. Tong (2000) finds pervasive weekday effects on 23 European, Asian and North American market indexes by considering bad-friday correlation and month factor. He documents that 15 of 22 non-us markets show the significant Monday effect except for Australia, Austria, Belgium, Indonesia, Taiwan and Japan. Aggarwal and Rivoli (1989) find weekday effects in four emerging Asian markets (Hong Kong, Singapore, Malaysia and the Philippines) from 1976 to 1988. They suggest that a strong negative Monday and Tuesday effect may be linked to the time differential between the location of these markets and the New York Stock Exchange. Wong, Hui and Chan (1992), investigating the period of 1975 to 1988 for Singapore, Malaysia, Hong Kong, Thailand, and Taiwan, find that the returns on all markets except for Taiwan are negative on Mondays or Tuesdays but highly positive on Fridays. However, Ajayi, Mehdian and Perry (2004), who investigate eleven Eastern European emerging markets (EEEM-Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Russia, Slovakia, and Slovenia) from the mid 1990s to 2002, find the negative Monday returns in six of the EEEMs and the positive Monday returns in the remaining five countries. Two of the six negative Monday returns and only one of the five positive Monday returns are statistically significant. These findings provide no consistent evidence supporting the presence of any significant daily patterns in the stock market returns of the EEEM. Hui (2005) examines the period of 1998 to 2001 for the US, Japan, Hong Kong, Korea, Singapore and Taiwan, and he finds no evidence of the day-of-the-week effect in all the countries except Singapore. In contrast, Cai, Li and Qi (2006) analyze Shanghai A, B share and Shenzhen A, B share indexes for the period of 1992 to 2002, and they conclude that average Monday returns on A- share indexes are significantly negative during the third and fourth weeks, as in the U.S. market. However, average Tuesday returns on most of the A-share and B-share indexes are negative during the second week of the month. Their results generally support the day-of-the week effect in the Chinese market. Moreover, Agathee (2008), using stock exchange index of Mauritius (SEMDEX) from 1998 to 2006, find that the Friday returns appeared to be higher relative to other trading days, and further empirical results suggest that the mean returns across the five week days are jointly not significantly different from zero across all given years as well as for the whole sample period. In contrast, Lim, Ho and Dollery (2010) examine day-of-the-week effect and the twist of the Monday effect for Kuala Lumpur Composite Index for the period 2000 to 2006. They document that Monday shows a negative mean return and represents the lowest stock returns in a week while the returns on Wednesday are the highest in a week, followed by returns on Friday. They also find that the Monday effect is clearly

81 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) visible in a bad news environment, but fail to appear in good news environment. They also find evidence on twist of the Monday effect, where returns on Mondays are influenced by previous week s returns and previous Friday s returns. Over the last three decades, several explanations on the day-of-the-week effect are proposed to discuss what factors might have contributed to the occurrence of negative equity returns on Mondays (See Pettengill, 2003), which are Monday Effect and Statistical Error, Micro Market Effects, Information Flow Effects, The Role of Order Flow, Conditional Nature of Monday s and Trading on the Monday. Each explanation attempts to show a possible reason that might create a certain pattern for the equity returns in term of day seasonality. Interestingly, some researchers provide some evidence to support their arguments. For instance, French (1980) suggests that the possible explanation of the weekday effect was a tendency for firms to delay the announcement of bad news until the weekend to avoid market disruption. Another example is that Monday returns are influenced by returns on the preceding trading days (e.g., Cross, 1973; Lim, Ho and Dollery, 2010). There are a few studies on the Australia and New Zealand stock markets such as Easton and Faff (1994), Davidson and Faff (1999), Keef and McGuinness (2001), Marrett and Worthington (2008) and Worthington (2010). Davidson and Faff (1999), using the Australian all ordinary accumulation index from 1983 to 1996, report that the day-of-the-effect (i.e. Tuesday effect) has disappeared in recent years. Marrett and Worthington (2008) examine the day-of-the-week effect in the Australian daily stock returns at the market and industry levels from 1996 to 2006. They find that the overall Australian market provides no evidence of daily seasonality but there is evidence of a small capitalization day-of-the-week effect with systematically higher returns on Thursdays and Fridays. Their analysis of the 10 sectors index returns is also partly supportive of the day-of-the-week effects in the banking, diversified financial, energy, healthcare, insurance, materials and retail industries. However, it hardly supports the lower Monday or Tuesday returns reported in earlier work. Furthermore, Worthington (2010), using the Australian daily stock index from 1958 through 2005, examines three calendar effects day-of-the-week, turn-of-the-month and month-of-the-year. He documents that the Australian market is characterized by the seasonality of all three forms with Tuesday, September and the second trading day of the month. 3. Data The data employed in this study are daily closing prices from the top 50 companies traded on the Australian Stock Exchange (ASX) over the period from January 2, 2001 to June 30, 2010. The prices are adjusted by dividend distribution, new equity issuance and share buyback. The data are sourced from DataStream. The top 50 companies are selected based on their market capitalization on June 30, 2010. One company, West Field American Trust (ASX code: WFA), is not included as the price series of this company is incorrect in DataStream. Therefore only 49 companies are used in this study. The detailed description of the companies and their associated industry category can be found in Table 1. Some companies started to list their stocks on the exchange after 2001, and therefore the sample starting period for these companies is different from January 2, 2001 (see Table 1 for the details). For example, the sample starting day for CWN is November 28, 2007. For most companies, the sample starting day is January 2, 2001, as shown in Table 1. The daily market return at day t is calculated as: Ri, t = ln( Pi, t / Pi, t 1), (1) where, i t P is the price of stock i at day t. Table 1 presents summary statistics of the daily returns. The sample means, standard deviations, medians, minimums, maximums, skewness, kurtosis, Jacaque-Bera statistics, and the firstorder autocorrelation coefficients are reported. The median returns for most companies are close to zero. Consistent with Marret and Worthington (2008), we find that the return distributions for most companies are non-normal. All Jarque-Bera statistics for normality test are significant at the 1% level,

82 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) suggesting the rejection of the null hypothesis. Furthermore, the kurtosis for most return series is significantly larger than 3, suggesting fat-tail distributions. Finally, the first-order autocorrelation coefficients for most companies are less than 0.1. Table 1: Summary Statistics ASX Std. Dev. Median Min Max Jarque Industry Skewness Kurtosis Code ( 100) ( 100) ( 100) ( 100) ( 100) -Bera ρ(1) Starting Day AGK Utilities 0.02 1.40 0.00-18.19 7.06-0.99 14.80 23001-0.02 Jan 2, 2001 AMP Insurance -0.04 2.16 0.00-44.39 20.96-3.25 77.87 630198 0.02 Jan 2, 2001 ANZ Banks 0.02 1.70 0.00-11.57 13.65 0.07 8.42 7327 0.03 Jan 2, 2001 ASX Diversified Financials 0.03 1.71 0.00-11.36 15.32 0.22 6.22 4007-0.02 Jan 2, 2001 AWC Materials -0.03 2.63 0.00-21.00 16.47-0.47 6.95 5072 0.06 Jan 2, 2001 AXA Insurance 0.02 2.34 0.00-24.24 33.65 1.58 33.26 115200-0.12 Jan 2, 2001 BHP Materials 0.06 2.12 0.04-14.08 11.46-0.21 3.70 1434-0.04 Jan 2, 2001 BSL Materials -0.01 2.60 0.00-30.16 20.16-0.97 13.06 15077 0.05 Jul 16, 2002 BXB Commercial Service -0.03 2.08 0.00-35.25 13.94-2.30 38.45 154777 0.02 Jan 2, 2001 CBA Banks 0.02 1.53 0.00-9.53 11.79 0.04 6.00 3715 0.01 Jan 2, 2001 CCL Food Beverage 0.04 1.61 0.00-11.91 12.09-0.07 4.25 1868-0.09 Jan 2, 2001 CNA Energy 0.07 1.83 0.00-16.33 20.66 1.26 26.39 72554-0.04 Jan 2, 2001 CPU Software & Services 0.01 2.48 0.00-41.64 22.31-1.65 37.98 149993 0.00 Jan 2, 2001 CSL Pharmaceuticals, Biotechnology & Life 0.04 2.08 0.00-12.02 18.89 0.20 6.18 3960 0.08 Jan 2, 2001 Sciences CWN Consumer Services -0.09 2.59 0.00-13.38 12.63 0.07 2.89 234-0.05 Nov 28, 2007 ERA Energy 0.06 3.06 0.00-26.02 17.91-0.42 6.85 4910 0.01 Jan 2, 2001 FGL Food Beverage 0.01 1.35 0.00-6.96 8.78 0.23 3.31 1151-0.03 Jan 2, 2001 FMG Materials 0.25 5.89 0.00-69.31 69.31 0.61 23.55 57383-0.05 Jan 2, 2001 GPT Real Estate -0.05 2.65 0.00-39.40 18.24-1.72 34.44 123658 0.12 Jan 2, 2001 IAG Insurance 0.01 1.74 0.00-16.15 11.57-0.48 6.35 4251-0.06 Jan 2, 2001 IPL Materials 0.08 2.74 0.00-36.34 20.58-1.47 23.98 43938 0.04 Jul 29, 2003 LEI Capital Goods 0.06 2.43 0.00-25.64 14.41-0.49 8.82 8134 0.02 Jan 2, 2001 LGL Materials 0.08 3.00 0.00-15.88 28.77 0.43 6.05 3856 0.00 Jan 2, 2001 MAP Transport 0.04 2.30 0.00-15.02 9.94-0.39 3.99 1417-0.02 Aug 13, 2002 MQG Diversified Financials 0.01 2.52 0.00-26.43 32.07 0.26 19.92 40971-0.01 Jan 2, 2001 NAB Banks -0.01 1.73 0.00-14.49 16.00-0.42 10.59 11649 0.03 Jan 2, 2001 NCM Materials 0.09 2.65 0.00-18.37 14.05-0.17 3.88 1569 0.03 Jan 2, 2001 NHC Energy 0.11 2.70 0.00-16.89 13.10 0.00 3.72 1021-0.02 Sep 17, 2003 ORG Energy 0.08 1.79 0.00-10.70 28.70 1.88 28.84 87326-0.06 Jan 2, 2001 ORI Materials 0.06 1.97 0.00-16.60 18.33 0.03 7.89 6426 0.02 Jan 2, 2001 OSH Energy 0.06 2.83 0.00-28.77 19.57-0.46 9.95 10305 0.01 Jan 2, 2001 QAN Transportation -0.02 2.10 0.00-20.26 23.43 0.10 14.58 21953-0.02 Jan 2, 2001 QBE Insurance 0.03 2.48 0.00-52.63 41.85-4.56 167.11 289066 9-0.10 Jan 2, 2001 RIO Materials 0.04 2.47 0.01-41.94 14.35-2.31 38.08 151907-0.02 Jan 2, 2001 SGM Materials 0.05 2.25 0.00-22.96 13.33-0.57 8.14 6965 0.00 Jan 2, 2001 SGP Real Estate 0.00 2.00 0.00-11.39 11.62-0.28 7.79 6303 0.07 Jan 2, 2001 SHL Health Care Equipment & Services 0.01 1.74 0.00-22.57 9.39-1.05 16.81 29605-0.06 Jan 2, 2001 STO Energy 0.03 2.03 0.00-16.84 11.34-0.29 5.51 3168 0.00 Jan 2, 2001 SUN Insurance -0.01 2.01 0.00-29.47 11.44-1.46 23.51 57937 0.03 Jan 2, 2001 TAH Consumer Services -0.02 1.51 0.00-23.95 9.07-1.83 28.85 87279-0.09 Jan 2, 2001 TCL Transportation 0.00 1.80 0.00-15.79 19.92 0.55 14.29 21202-0.06 Jan 2, 2001 TLS Telecommunication Service -0.03 1.36 0.00-12.36 6.47-0.66 5.82 3670 0.02 Jan 2, 2001 TOL Transportation 0.05 2.08 0.00-19.75 12.91-0.16 8.56 7579 0.05 Jan 2, 2001 WBC Banks 0.02 1.59 0.00-11.82 8.59-0.08 5.00 2583 0.00 Jan 2, 2001 WDC Real Estate 0.00 1.83 0.00-13.24 20.92 0.53 11.71 14263-0.05 Jan 2, 2001 WES Food & Stapling Retailing 0.03 1.80 0.00-14.40 12.59-0.44 8.78 8030 0.01 Jan 2, 2001 WOR Energy 0.13 2.66 0.00-16.27 20.74 0.21 6.82 3855 0.08 Nov 29, 2002 WOW Food & Stapling Retailing 0.05 1.33 0.00-11.33 6.38-0.28 4.61 2229-0.04 Jan 2, 2001 WPL Energy 0.04 1.93 0.00-11.95 12.08-0.13 4.57 2161 0.02 Jan 2, 2001 Notes: The firms are: AGK-AGL Energy, AMP, ANZ-ANZ Bank, ASX, AWC-Aluminum Limited, AXA-AXA Asia, BHP-BHP BLT, BSL-Bluescope, BXB- Bramble Ltd, CBA-Commonwealth Bank, CCL-Coca Cola Amatil, CNA- Coal & Allied Industries, CPU-Cshare, CSL, CWN-Crown, ERA-Energy Resources of Australia, FGL-Fosters, FMG-Fortescue, GPT, IAG-Insurance Australia, IPL-Incitec PV, LEI-Leighton, LGL-Lihir., MAP-Airport Investment Fund, MQG-Macquarie Group, NAB-National Australian Bank, NCM-Newcrest, NHC-New Hope Corporation, ORG-Origin Energy, ORI-Orica, OSH-Oil Search, QAN-Qantas, QBE-QBE Insurance, SGM-Sims Metal Management, RIO-Rio Tinto, SGP-Stockland, SHL-Sonic Health, Sun-Sun Metway, TAH-Tabcorp, TCL-

83 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) Transurban, TLS-Telstra, TOL-Tollholdings, WBC-Westpac, WDC-Westfield, WES-Wesfamrer, WOR- Worleypars, WOW-Woolworths, WPL-Woodside. WFA-Westfield American Trust is not included as its stock price data are invalid in the database. All Jarque-Bera statistics for normality are significant at the 1% level. The samples are daily and end on June 30, 2010. 4. Empirical Approaches and Results We use usual t-tests to test the day-of-the-week hypothesis. Following Marret and Worthington (2008), we investigate the day-of-the-week effect on the basis of a trading day hypothesis whereby returns are calculated on trading days during the week. To be specific, we calculate mean return on each weekday (Monday to Friday), and mean return on other four weekdays. Then we calculate the difference of mean returns and use t-tests to test the statistical significance of test return. For example, to test the Monday effect, the t-statistic is calculated as follows: RMon RNonMon t =, (2) 2 2 SMon SNonMon + n n where Monday, Mon NonMon RMon is the mean return on Monday, RNonMon S 2 Mon nmon and nnonmon is the variance of Monday returns, S is the mean return on the weekdays other than 2 NonMon is the variance of Non-Mondays returns, and are the observation numbers of Monday returns and Non-Mondays returns, respectively. Before the t-tests, we present the mean returns of 49 companies on each weekday (from Monday to Friday) and their associated standard errors of mean in Table 2. We also report the largest return day and the lowest return day for each company in the rightmost two columns of Table 2. In contrast to the US studies, which find returns on Mondays are much lower than on the other four days, Table 2 shows that 13 out of 49 companies have significant positive returns on Monday in the Australian market. However the large positive Monday returns are concentrated in the materials and energy companies. Though there are several other-industries companies displaying positive returns on Monday such as WES and WOW, the magnitude of their mean returns are far less than the magnitude of the materials and energy companies. The early studies on the Australian stock market such as Ball and Bowers (1988), and Finn, Lynch, and Moore (1991) find that the large negative return occurs on Tuesday. However, we find different results. Table 2 shows that only three companies (BSL, LGL and SHL) have statistically significant negative returns on Tuesdays. Moreover, three companies (AGK, ORI, and WBC) have statistically positive mean returns on Tuesdays. The mean returns for all the companies on Wednesdays are not significantly different from zero except for CSL. Five companies display positive mean returns on Thursdays. Five companies (BXB, GPT, NCM, ORI, and WDC) have negative mean returns and four companies (LGL, NCM, NHC, QBE) have positive mean returns on Fridays. We also find that the largest mean weekday returns occur on Monday for 15 companies, and lowest mean weekday returns occur on Friday for 15 companies. Table 3 reports the t-testing results of equation (2) for 49 companies. Consistent with Table 2, Table 3 shows that returns on Monday are significant larger than the other four days for six companies. These companies are the materials and energy companies. Moreover, the magnitude of the difference is quite large. For example, for LGL, the mean difference is 0.574% per day, which amounts to 144.65% on an annualized basis. There is no strong evidence of other-than-monday-of-the-week effect in the sample. There are two companies (LGL and SHL) whose Tuesday returns are significantly less than the returns on the other four days. RIO has significantly lower return on Wednesday than the returns on other four days, and ORI has significantly lower return on Friday than the returns on other four days. However, two companies (FGL and OSH) have significantly higher returns on Thursday than on other weekdays.

84 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) Table 2: s on Weekdays ASX Code Monday Tuesday Wednesday Thursday Friday Error of Error of Error of Error of Error of Largest Day Smallest Day AGK -0.031 (0.066) 0.123** (0.057) 0.009 (0.061) 0.030 (0.067) -0.032 (0.064) Tue Fri AMP -0.111 (0.127) -0.109 (0.085) -0.043 (0.082) 0.032 (0.096) 0.035 (0.087) Fri Mon ANZ 0.009 (0.081) 0.081 (0.073) 0.069 (0.075) -0.033 (0.076) -0.035 (0.077) Tue Fri ASX -0.011 (0.076) 0.099 (0.076) 0.008 (0.077) 0.097 (0.083) -0.022 (0.070) Tue Fri AWC 0.159 (0.113) -0.163 (0.116) -0.114 (0.128) -0.011 (0.122) -0.032 (0.110) Mon Tue AXA -0.057 (0.125) 0.053 (0.088) 0.036 (0.094) -0.007 (0.094) 0.097 (0.119) Fri Mon BHP 0.277** (0.094) 0.019 (0.094) 0.013 (0.094) -0.008 (0.105) -0.005 (0.088) Mon Thu BSL 0.208* (0.121) -0.202* (0.122) -0.133 (0.123) 0.093 (0.155) 0.003 (0.114) Mon Tue BXB 0.014 (0.080) -0.007 (0.093) 0.071 (0.095) -0.056 (0.114) -0.147* (0.081) Wed Fri CBA -0.037 (0.073) 0.049 (0.072) 0.014 (0.067) 0.037 (0.067) 0.030 (0.063) Tue Mon CCL 0.034 (0.075) -0.039 (0.070) 0.033 (0.073) 0.098 (0.078) 0.079 (0.065) Thu Tue CNA 0.097 (0.085) 0.044 (0.077) 0.046 (0.078) 0.047 (0.086) 0.092 (0.084) Mon Tue CPU 0.012 (0.094) -0.019 (0.095) -0.037 (0.111) -0.048 (0.144) 0.134 (0.105) Fri Thu CSL 0.003 (0.088) -0.106 (0.090) 0.203** (0.102) 0.030 (0.096) 0.055 (0.090) Wed Tue CWN -0.286 (0.220) -0.260 (0.210) -0.016 (0.214) -0.103 (0.230) 0.213 (0.243) Fri Mon ERA 0.203 (0.125) 0.038 (0.134) -0.061 (0.147) 0.026 (0.139) 0.109 (0.141) Mon Wed FGL 0.018 (0.055) -0.046 (0.060) -0.045 (0.066) 0.150** (0.061) -0.041 (0.061) Thu Tue FMG 0.782** (0.240) -0.051 (0.291) 0.094 (0.262) 0.295 (0.276) 0.141 (0.250) Mon Tue GPT -0.125 (0.116) 0.026 (0.122) -0.052 (0.113) 0.149 (0.108) -0.259* (0.134) Thu Fri IAG -0.046 (0.081) 0.034 (0.073) 0.086 (0.079) 0.067 (0.079) -0.097 (0.079) Wed Fri IPL 0.099 (0.129) -0.085 (0.170) 0.109 (0.152) 0.118 (0.132) 0.138 (0.134) Fri Tue LEI 0.250** (0.114) 0.002 (0.112) -0.054 (0.101) 0.051 (0.117) 0.061 (0.101) Mon Wed LGL 0.541** (0.133) -0.398** (0.124) 0.042 (0.130) 0.018 (0.159) 0.209* (0.123) Mon Tue MAP -0.032 (0.103) 0.122 (0.114) -0.075 (0.114) 0.197* (0.119) -0.034 (0.117) Thu Wed MQG -0.079 (0.107) 0.067 (0.115) -0.006 (0.114) 0.066 (0.114) 0.003 (0.117) Tue Mon NAB -0.015 (0.079) 0.090 (0.074) 0.052 (0.073) -0.020 (0.073) -0.151* (0.090) Tue Fri NCM 0.405** (0.119) -0.078 (0.110) -0.089 (0.121) -0.037 (0.132) 0.237** (0.110) Mon Wed NHC 0.144 (0.141) -0.105 (0.142) -0.057 (0.138) 0.281* (0.150) 0.280* (0.147) Thu Tue ORG 0.038 (0.076) 0.088 (0.077) 0.051 (0.097) 0.184** (0.073) 0.046 (0.076) Thu Mon ORI 0.024 (0.094) 0.175** (0.087) 0.122 (0.096) 0.135 (0.086) -0.148* (0.077) Tue Fri OSH 0.246** (0.115) -0.137 (0.131) -0.006 (0.137) 0.303** (0.132) -0.124 (0.118) Thu Tue QAN -0.030 (0.086) -0.007 (0.099) -0.012 (0.094) -0.041 (0.106) -0.003 (0.085) Fri Thu QBE -0.147 (0.125) -0.020 (0.084) 0.025 (0.085) 0.016 (0.134) 0.252** (0.118) Fri Mon RIO 0.249** (0.103) 0.063 (0.100) -0.186 (0.131) 0.016 (0.120) 0.070 (0.097) Mon Wed SGM 0.128 (0.095) -0.019 (0.092) -0.064 (0.097) 0.181* (0.109) 0.005 (0.110) Thu Wed SGP 0.005 (0.084) 0.052 (0.093) -0.022 (0.090) 0.049 (0.090) -0.082 (0.092) Tue Fri SHL 0.039 (0.073) -0.158** (0.080) 0.048 (0.080) 0.020 (0.072) 0.087 (0.086) Fri Tue STO 0.191** (0.087) -0.065 (0.085) -0.076 (0.090) 0.100 (0.101) 0.016 (0.092) Mon Wed SUN -0.084 (0.107) -0.065 (0.083) 0.021 (0.085) 0.062 (0.086) 0.021 (0.088) Thu Mon TAH -0.028 (0.062) -0.046 (0.067) 0.001 (0.062) 0.013 (0.070) -0.051 (0.077) Thu Fri TCL -0.099 (0.085) 0.005 (0.080) 0.045 (0.074) 0.127 (0.081) -0.090 (0.083) Thu Mon TLS -0.075 (0.067) -0.033 (0.060) 0.011 (0.061) -0.009 (0.060) -0.032 (0.057) Wed Mon TOL 0.115 (0.096) -0.044 (0.089) 0.121 (0.086) 0.129 (0.109) -0.054 (0.085) Thu Fri WBC -0.015 (0.073) 0.123* (0.069) -0.001 (0.074) 0.010 (0.074) -0.021 (0.067) Tue Fri WDC 0.000 (0.072) 0.066 (0.091) -0.008 (0.084) 0.061 (0.081) -0.137* (0.083) Tue Fri WES 0.129* (0.078) -0.049 (0.085) 0.098 (0.079) 0.008 (0.081) -0.030 (0.080) Mon Tue WOR 0.249* (0.139) 0.132 (0.134) 0.154 (0.129) 0.088 (0.141) 0.028 (0.126) Mon Fri WOW 0.142** (0.062) 0.067 (0.062) -0.004 (0.059) 0.057 (0.055) -0.027 (0.062) Mon Fri WPL 0.204** (0.086) 0.001 (0.090) -0.030 (0.079) -0.002 (0.096) 0.039 (0.082) Mon Wed Notes: returns and their associated standard errors of mean are expressed in percentages. returns which are statistically significant different from zero at the 5% and 10% levels are denoted with ** and *, respectively. The samples are daily, and start from January 2, 2001 for most firms, and end on June 30, 2010 for all firms.

85 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) Table 3: Test of Difference ASX Code Monday-Non Monday Differenc e Error Tuesday-Non Tuesday Differenc e Error Wednesday-Non Wednesday Differenc e Error Thursday-Non Thursday Differenc e Error Friday-Non Friday Differenc e Error AGK -0.063 (0.091) 0.129 (0.086) -0.014 (0.088) 0.012 (0.091) -0.065 (0.090) AMP -0.090 (0.155) -0.087 (0.131) -0.005 (0.129) 0.089 (0.137) 0.092 (0.132) ANZ -0.011 (0.110) 0.078 (0.106) 0.063 (0.107) -0.064 (0.108) -0.067 (0.109) ASX -0.057 (0.108) 0.081 (0.108) -0.033 (0.108) 0.079 (0.112) -0.070 (0.105) AWC 0.239 (0.164) -0.164 (0.166) -0.102 (0.172) 0.027 (0.169) 0.001 (0.163) AXA -0.102 (0.160) 0.036 (0.140) 0.015 (0.143) -0.040 (0.143) 0.090 (0.157) BHP 0.272** (0.134) -0.050 (0.134) -0.057 (0.134) -0.084 (0.140) -0.081 (0.131) BSL 0.268 (0.177) -0.245 (0.178) -0.158 (0.178) 0.125 (0.196) 0.012 (0.174) BXB 0.049 (0.125) 0.022 (0.132) 0.121 (0.133) -0.039 (0.144) -0.153 (0.126) CBA -0.070 (0.100) 0.038 (0.099) -0.006 (0.096) 0.023 (0.096) 0.014 (0.094) CCL -0.009 (0.104) -0.100 (0.101) -0.010 (0.103) 0.071 (0.106) 0.047 (0.099) CNA 0.040 (0.118) -0.027 (0.114) -0.024 (0.114) -0.023 (0.118) 0.033 (0.118) CPU 0.005 (0.149) -0.034 (0.149) -0.057 (0.157) -0.070 (0.176) 0.156 (0.154) CSL -0.043 (0.129) -0.179 (0.130) 0.208 (0.137) -0.008 (0.134) 0.022 (0.130) CWN -0.244 (0.314) -0.212 (0.309) 0.093 (0.311) -0.015 (0.320) 0.379 (0.326) ERA 0.175 (0.188) -0.031 (0.192) -0.155 (0.200) -0.046 (0.195) 0.057 (0.196) FGL 0.014 (0.083) -0.066 (0.086) -0.065 (0.089) 0.178** (0.086) -0.060 (0.086) FMG 0.663* (0.362) -0.379 (0.389) -0.198 (0.373) 0.053 (0.381) -0.139 (0.367) GPT -0.091 (0.167) 0.097 (0.170) 0.001 (0.165) 0.252 (0.162) -0.259 (0.177) IAG -0.069 (0.112) 0.031 (0.108) 0.096 (0.111) 0.073 (0.111) -0.132 (0.111) IPL 0.029 (0.196) -0.202 (0.218) 0.042 (0.208) 0.053 (0.197) 0.078 (0.198) LEI 0.235 (0.157) -0.074 (0.156) -0.146 (0.150) -0.013 (0.159) -0.001 (0.151) LGL 0.574** (0.190) -0.600** (0.185) -0.051 (0.188) -0.080 (0.204) 0.158 (0.185) MAP -0.084 (0.155) 0.108 (0.161) -0.139 (0.161) 0.202 (0.163) -0.086 (0.163) MQG -0.111 (0.157) 0.071 (0.161) -0.020 (0.160) 0.069 (0.160) -0.009 (0.162) NAB -0.008 (0.111) 0.124 (0.108) 0.076 (0.108) -0.014 (0.108) -0.178 (0.117) NCM 0.396** (0.168) -0.207 (0.164) -0.221 (0.170) -0.155 (0.176) 0.187 (0.164) NHC 0.044 (0.202) -0.267 (0.202) -0.207 (0.200) 0.216 (0.206) 0.215 (0.205) ORG -0.054 (0.112) 0.009 (0.112) -0.038 (0.123) 0.128 (0.110) -0.044 (0.111) ORI -0.047 (0.128) 0.142 (0.124) 0.076 (0.129) 0.091 (0.124) -0.262** (0.119) OSH 0.237 (0.174) -0.242 (0.182) -0.078 (0.185) 0.308* (0.182) -0.225 (0.175) QAN -0.014 (0.129) 0.014 (0.136) 0.008 (0.133) -0.028 (0.140) 0.019 (0.129) QBE -0.215 (0.165) -0.057 (0.144) 0.000 (0.144) -0.012 (0.170) 0.284* (0.161) RIO 0.258* (0.153) 0.026 (0.151) -0.285* (0.168) -0.033 (0.162) 0.035 (0.150) SGM 0.102 (0.139) -0.081 (0.138) -0.138 (0.141) 0.168 (0.147) -0.051 (0.148) SGP 0.005 (0.124) 0.065 (0.128) -0.028 (0.127) 0.061 (0.127) -0.104 (0.128) SHL 0.040 (0.108) -0.207* (0.111) 0.051 (0.112) 0.016 (0.108) 0.100 (0.115) STO 0.197 (0.127) -0.123 (0.126) -0.137 (0.129) 0.084 (0.135) -0.021 (0.130) SUN -0.094 (0.137) -0.069 (0.124) 0.037 (0.125) 0.089 (0.125) 0.038 (0.126) TAH -0.007 (0.093) -0.030 (0.095) 0.029 (0.093) 0.043 (0.097) -0.036 (0.101) TCL -0.120 (0.116) 0.009 (0.114) 0.060 (0.111) 0.161 (0.115) -0.110 (0.115) TLS -0.059 (0.089) -0.007 (0.086) 0.049 (0.086) 0.023 (0.086) -0.005 (0.084) TOL 0.077 (0.134) -0.122 (0.130) 0.084 (0.128) 0.095 (0.141) -0.135 (0.128) WBC -0.043 (0.102) 0.130 (0.100) -0.025 (0.102) -0.012 (0.102) -0.050 (0.099) WDC 0.004 (0.111) 0.087 (0.121) -0.005 (0.118) 0.080 (0.116) -0.167 (0.117) WES 0.122 (0.113) -0.100 (0.116) 0.083 (0.113) -0.029 (0.114) -0.076 (0.114) WOR 0.148 (0.192) 0.002 (0.190) 0.030 (0.187) -0.053 (0.193) -0.127 (0.185) WOW 0.119 (0.086) 0.025 (0.086) -0.064 (0.084) 0.013 (0.082) -0.093 (0.086) WPL 0.202* (0.122) -0.052 (0.124) -0.090 (0.119) -0.055 (0.128) -0.005 (0.120) Notes: differences and their associated standard errors are expressed in percentages. differences which are statistically significant different from zero at the 5% and 10% levels are denoted with ** and *, respectively. The samples are daily, and start from January 2, 2001 for most firms, and end on June 30, 2010 for all firms. Mehdian and Perry (2001) find that the October 1987 market crash affects the day-of-the-week pattern in the US stock market. The Global Financial Crisis, which starts from 2007, might affect the day-of-the-week pattern in the Australian stock market. To perform a robustness analysis, we perform another test using the sample ending on December 31, 2006. The results (not reported in tables to save space) are similar to Table 3. We still find that the materials and energy companies generally have higher returns on Mondays than other weekdays.

86 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) 5. Conclusion Seasonal or calendar anomalies in equity markets (over specific days, weeks, months, and even years) have brought a widespread attention and considerable interests. During the last three decades, many researchers have documented evidence on day seasonality of stock markets around the world. The research includes those of Cross (1973), French (1980), Gibbons and Hess (1981) and Linton and Whang (2007), who all report significantly negative average returns on the US stock markets on Mondays. Similar anomalies are found in international equity markets as well. However, some recent literature reports the declining or reversal weekend effects and a different shift of the weekend effect on large capitalization securities and small capitalization securities. Findings of this topic on the Australian equity market are mixed, depending on the sample period and the portfolios used. However, all the studies are only limited to the use of portfolio data and none of them use individual stock data. As stock returns of different companies may have different day anomalies, to generate new findings we investigate Australian equity seasonality using the top 50 companies stocks for the period of January 2001 through June 2010. We find that the largest mean weekday returns occur on Monday for 15 companies, and lowest mean weekday returns occur on Friday for 15 companies. Findings are mixed across companies and industries regarding the weekday effects. Further tests indicate that returns on Monday are significant larger than that on the other four days for six companies which are the materials and energy companies. In other words, these material and energy companies demonstrate positive anomalies on Mondays, rather than on Tuesdays in literature (e.g., Dubois and Louvet, 1996). The results lend some support to the reverse weekend effects (e.g., Connolly, 1989; Kamara, 1997; Doyle and Chen, 2007). In addition, the magnitude of the difference between returns on Mondays and Non-Mondays are quite large. For example, for LGL, the mean difference is 0.574% per day, which amounts to 144.65% on an annualized basis. The results further indicated that there is no strong evidence of other-than-mondayof-the-week effect in the sample. As this study is limited to the top 50 companies, further research may be needed to consider more companies for a longer time period. References 1] Agathee, U.S., 2008. Day of the Week Effects: Evidence from the Stock Exchange of Mauritius (SEM), International Research Journal of Finance and Economics 17, pp. 7-14. 2] Aggarwal, R., and P. Rivoli, 1989. Seasonal and Day-of-the-Week Effects in Four Emerging Stock Markets, Financial Review 24, pp. 541-550. 3] Ajayi, R.A., S. Mehdian and M.J. Perry, 2004. The Day-of-the-Week Effect in Stock s: Further Evidence from Eastern European Emerging Markets, Emerging Markets Finance and Trade 40, pp. 53-62. 4] Ball, R., and J. Bowers, 1988. Daily Seasonal in Equity and Fixed Interest s Australian Evidence and Tests of Plausible Hypothesis, in E. Dimson (ed.), Stock Market Anomalies, Cambridge University Press, Cambridge. 5] Bessembinder, H., and M. Hertzel, 1993. Autocorrelation around Nontrading Days, Review of Financial Studies 6, pp. 155-189. 6] Chang, E., J. Pinegar, and R. Ravichandran, 1993. International Evidence on the Robustness of the Day-of-the-Week Effect, Journal of Financial and Quantitative Analysis 28, pp. 497-513. 7] Connolly, R., 1989. An Examination of the Robustness of the Weekend Effect, Journal of Financial and Quantitative Analysis 24, pp. 133-170. 8] Cross, F., 1973. The Behavior of Stock Prices on Fridays and Mondays, Financial Analysts Journal 29, pp. 67-69. 9] Cai, J., Y. Li and Y. Qi, 2006. The Day-of-the-Week Effect: New Evidence from the Chinese Stock Market, The Chinese Economy 39, pp. 71 88.

87 European Journal of Economics, Finance and Administrative Sciences - Issue 24 (2010) 10] Davidson, S. and R. W. Faff, 1999. Some Additional Australian Evidence on the Day-of-the- Week Effect, Applied Economics Letters 6, pp. 247-249. 11] Doyle, J.R. and C.H. Chen, 2007. The Wandering Weekday Effect in Major Stock Markets, Journal of Banking and Finance 33, pp. 1388-1399. 12] Easton, S.A. and R.W. Faff, 1994. An Investigation of the Robustness of the Day-of-the-Week Effect in Australia, Applied Financial Economics 4, pp. 99-110. 13] French, K.R., 1980. Stock s and the Weekend Effect, Journal of Financial Economics 8, pp. 55 69. 14] Gardeazabal, J., and M. Regulez, 2004. A Factor Model of Seasonality in Stock s, The Quarterly Review of Economics and Finance 44, pp. 224-236. 15] Gibbons, M., and P. Hess, 1981. Day of the Week Effects and Asset s, Journal of Business 54, pp. 579-596. 16] Gu, A.Y., 2004. The Reversing Weekend Effect: Evidence from the U.S. Equity Market, Review of Quantitative Finance and Accounting 22, pp. 5-14. 17] Hui, T.K., 2005. Day-of-the-Week Effects in the US and Asia-Pacific Stock Markets during the Asian Financial Crisis: A Non-parametric Approach, Omega 33, pp. 277-282. 18] Hindmarch, S., D. Jentsch, and D. Drew, 1984. A Note on Canadian Stock s and the Weekend Effect, Journal of Business Administration 14, pp. 163-172. 19] Fama, E.F., 1965. The Behavior of Stock Market Prices, Journal of Business 38, pp. 34 105. 20] Finn, F.J., A. Lynch and S. Moore, 1991. Intra-week Regularities in Security s: Further Australian Evidence, Australian Journal of Management 16, pp. 129-144. 21] Jaffe, J., and R. Westerfield, 1985. The Weekend Effect in Common Stock s: The International Evidence, Journal of Finance 40, pp. 433-454. 22] Keim, D., and R. Stambaugh, 1984. A Further Investigation of the Weekend Effect in Stock s, Journal of Finance 39, pp. 819-835. 23] Keef, S., and P. McGuinness, 2001. Changes in Settlement Regime and the Modulation of Day-of-the-Week Effects in Stock s, Applied Financial Economics 11, pp. 361-371. 24] Lim, S. Y., C. M. Ho and B. Dollery, 2010. An Empirical Analysis of Calendar Anomalies in the Malaysian Stock Market, Applied Financial Economics 20, pp. 255 264. 25] Linn, S., and L. Lockwood, 1988. Short-Term Stock Price Patterns: NYSE, AMEX, OTC, Journal of Portfolio Management 14, pp. 30-34. 26] Marrett, G.E., and A.C. Worthington, 2008. The Day-of-the-Week Effect in the Australian Stock Market: An Empirical Note on the Market, Industry and Small cap Effects, International Journal of Business and Management 3, pp. 3-8. 27] Mehdian, S., and M. Perry, 2001. The Reversal of the Monday Effect: New Evidence from US Equity Markets, Journal of Business Finance and Accounting 28, pp. 1043-1065. 28] Pettengill, G.N., 2003. A Survey of the Monday Effect Literature, Quarterly Journal of Business and Economics 42, pp. 121 137. 29] Sullivan, J.H., 2004. Market Breadth and the Monday Seasonal in Stock s, Quarterly Journal of Business and Economics 42, pp. 65-72. 30] Wong, K.A., T.K. Hui and C.Y. Chan, 1992. Day-of-Week Effects: Evidence from Developing Stock Markets, Applied Financial Economics 2, pp. 49-56. 31] Worthington, A.C., 2010. The Decline of Calendar Seasonality in the Australian Stock Exchange, 1958-2005, Annals of Finance 6, pp. 421-433.