Day-of-the-week effect on the return and conditional variance of the H-shares index in Hong Kong

Abstract

The purpose of this article is to investigate the day-of-the-week effect on both the return and conditional variance (volatility) of the H-shares index in Hong Kong from 3 January 2000 to 1 August 2008. Using an Exponential General Autoregressive Conditional Heteroskedasticity (EGARCH) specification to model the conditional variance, we find that the day-of-the-week effect is present in both return and variance equations. In particular, higher risk-adjusted returns are found on Monday and Friday. However, after adjusting for market risks that vary across the days of the week, only the Monday effect remains. The conditional variance model also finds that the highest volatility of return also occurs on Monday. Thus, the Monday effects on risk-adjusted returns may be a reward for higher volatility on that day. However, after adjusting for transaction costs, the abnormal returns for Monday become negligible.

Keywords:

• EGARCH model
• volatility
• H-shares index
• day-of-the-week effect

JEL Classification:

• G12
• G14
• G15

Notes

¹The H-shares index (Hang Seng China Enterprises index) was launched in 1994 in order to measure the performance of H-shares. The details of the H shares and the H-shares index are described on the website of the Hang Seng Indexes Co. Ltd. (www.hsi.com.hk).

²While the conditional variance is modelled as an Exponential General Autoregressive Conditional Heteroskedasticity (EGARCH) specification, Model 1 is known as an EGARCH-M model. Similarly, Clare et al. (1998) and Kiymaz and Berument (2003) applied the GARCH-M model to the investigation of the day-of-the-week effect in some stock market indices.

³The Shanghai A-share index tracks the price performance of all A-shares listed on the Shanghai Stock Exchange in China (www.sse.com.cn).

⁴Our results of significant positive Monday effects on H-shares index returns are in contrast to those studies using the Hang Seng index as a case study of the Hong Kong stock market; those studies provide evidence of negative Monday effects. See Lee et al. (1990), Ho (1990), Wong et al. (1992) and Agrawal and Tandon (1994).

⁵See the HKEx Fact Book 2008, Hong Kong Exchanges and Clearing Ltd.

⁶Likewise, settlement procedures cannot explain the daily seasonal anomaly in stock markets of many western countries (Agrawal and Tandon, 1994) and Mainland China (Wong et al., 1999).

⁷Effective 1 April 2003, the brokerage fee is freely negotiable between brokers and their clients. For example, the Hong Kong and Shanghai Banking Corporation (HSBC) charges its clients 0.25% (www.hsbc.com.hk). Effective 1 October 2010, the transaction levy is reduced from 0.004% to 0.003%. The details of transaction costs can be found in Transaction costs, Hong Kong Exchanges and Clearing Ltd. (www.hkex.com.hk).