Abstract
The paper uses rolling sample tests to investigate time-varying calendar effects in the Chinese stock market, based on the GARCH (1, 1)-GED model. The Friday effect existed with low volatility at the early stage, but it seems to have disappeared since 1997. The positive Tuesday effect began to appear then. There is a small-firm January effect with high volatility. The turn-of-the month effect has also disappeared in the Chinese stock market since 1997.
Keywords:
Acknowledgement
The research was financially supported by the “Study on the Evolution of Complex Economic System” at the Innovation Center of Economic Transition and Development of the Ministry of Education, People's Republic of China. Our special thanks are given to Professor Xiaoming Li for his excellent research assistance.
Notes
1To our best knowledge, no other researchers have ever used rolling sample tests in examining the calendar effect.
2If 1995–1996 samples begin from 16 January, only two weeks later than 1 January 1995, we can get a positive Friday effect at 5%. So, the Friday effect is particularly dependent on samples.