Does the weekday effect of the yen/dollar spot rates exist in Tokyo, London, and New York? An analysis of panel probability distribution

Abstract

Using probability distribution techniques, this article explores whether any differences exist between the returns and volatility of yen/dollar spot markets in Tokyo, London and New York. After the intraday returns were fit into probability distributions, New York is found to have the highest return, followed by London, and then Tokyo. In estimating the peaks and widths of the distributions of volatility, Tokyo is found to have the lowest volatility in the log-normal distribution, while London and New York show similar volatility distributions, implying similar investor risk-return preference behaviour in the London and New York markets. The findings also imply that arbitrage opportunities between London and New York could be trivial. After estimating the panel distribution from Monday to Friday across the three markets, we found that the Monday effect disappears. Instead, Tuesday shows negative and significantly lower returns. The Friday (weekend) effect no longer exists. Instead, Thursday shows a positive and significantly higher return than the other weekdays. Finally, the three major currency markets did not become more volatile after Japan's deregulation in the foreign currency market in April 1998. On the contrary, they show less volatile behaviour than before deregulation. The probability distributions of volatility on different weekdays did not change significantly after deregulation.

¹ This article had been presented in 13th Conference on Panel Data at the University of Cambridge 2006, and The Joint 14th Annual PBFEA and 2006 Annual FeAT Conference at Taipei.

Acknowledgements

We appreciate the helpful and suggestive comments from the refrees and editor Mark Taylor. We express our thanks to Albert G. Crawford, fot the editing.

Notes

¹ This article had been presented in 13th Conference on Panel Data at the University of Cambridge 2006, and The Joint 14th Annual PBFEA and 2006 Annual FeAT Conference at Taipei.

² Prior to 1 October 1981, all purchases of non-US dollar currencies were settled in two business days, whereas purchases of US dollars were settled in three days. This led to a relatively low demand for US dollars on Wednesday to avoid the 5-day interest rate charged by sellers. Hence, the significant Wednesday effect was present. Since 1981, both of purchases and sales have been treated the same in settlements. Therefore, this calendar effect may have been eliminated.