earticle

영어

The purpose of this study is to verify the effect of stock market floating policy that took place in the past and to present an empirical basis about the necessity of floating policy and the effect of policy. The researcher analyzed empirically the stock market floating policy conducted 5 times since 1995 with the data of daily general stock price index and transaction volume, in order to clarify how much effect does the government intervention have on the stock price and the reverse turn of earning rate, and what change does it occur onto the pattern of investment. It was confirmed that government`s frequent and successive market intervention has a little effect on the entire trend of the stock market. Generally speaking, the government`s floating policy conducted 5 times since 1995 seems to have only a short-term effect on the market. A temporal reverse turn occurs just after enforcing a policy, but it doesn`t last for 5 days, hardly having an effect on the fluctuation of the earning rate. According to the fluctuation of cumulative net purchase volume and net purchase sum by investors. the stock authorities` floating policy seemed to have an effect on the regular fluctuation of net purchase volume and net purchase sum. However, the 5 times of floating policy since 1995 had an effect on investors differently according to the contents. Individual or foreign investors showed frequent change in both net purchase volume and sum. but institutional investors didn`t show a remarkable change. An ARIMA(0,1,0) model was built up to verify the effectiveness of stock market floating policy statistically. After presuming the effectiveness during 40 days before the policy and 80 days before and after the policy, t-value was not significant statistically in both cases. So it isn`t correct statistically to say that stock authorities` floating policy has a positive effect on the earning rate. But in the t-test for during 40 days before the policy and 40 days after the policy by using the cumulative rate of return, we found that the difference of mean and standard deviation between two period was proved statistically in the 95% confidence level.