Estimating the Leverage Effect Using Panel Data With a Large Number of Stock Issues Over Short-Run Focus on the Tokyo Stock Exchange

This paper explores the leverage effect (the negative association between the stock return today and the stock return's volatility tomorrow), by utilizing the exponential ARCH type specification for panel data with a large number of stock issues and a small number of daily time series observations. The leverage effect is significantly estimated in the span from June 22 to 29 in 1998. This is the period right after the joint intervention into the yen-dollar exchange market. Judging from this fact, we may say that the leverage effect is associated with the deregulations in Japanese market incidental to the intervention