추종 거래와 주가 시계열 : ARMA 계수의 경제적 의미

In this study, we performed theoretical and empirical analysis on the characteristics of stock price time series and the stock return volatility in the presence of mimicking trader. The mimicking trade, which is followed by the trade of informed traders, has existed for a long time in the stock market. Moreover, as information communication technology is improved and its infrastructure expands, the speed of information spillover has accelerated and the mimicking lag has been shortened. This study is based on the idea that ‘mimicking trade’ may cause the stock price time series to follow the autoregressive moving average model (ARMA). To do this analysis, we classified the stock market participants into three groups, informed trader, mimicking trader and noise trader, and modeled buying and selling activities of three groups. This study shows that the stock return in equilibrium converges on the ARMA(1,1) process when ‘mimicking trade’ is introduced into the model, and we put an economic meaning on the ARMA coefficient by inducing the informed trader ratio through the coefficient. This means that the percentage of informed trader in total market participants can be estimated by ARMA(1,1) coefficient. On the other hand, we also studied the relationship between the market participant composition and the stock return volatility. In this study, the higher the mimicking trader ratio, the higher the stock return volatility is. However, the increase of the informed trader ratio does not always make the stock return volatility increased.