The purpose of this study is to construct composite information by considering feedback effect, such as information spillover from the US market to the domestic market, and to analyze that the effect is an asymmetric function of the domestic KOSPI index return. This study uses an extended partial adjustment model to examine whether the stock return and volatility shown in the KOSPI index return show different adjustment processes in the shock pattern of past information. To this purpose, the price adjustment model proposed by Koutmos (1998) is examined, and based on this, the model including the external impact filter in the study of Chen, Chiang, and So (2007) is partially applied and used for analysis. The reaction of previous day's composite information shows a significant negative value in AR(1) term when negative news is provided, and it appears insignificant when positive information is provided. This indicates that feedback trading exists in the domestic stock market, and positive feedback trading dominates when negative news is provided the previous day. The meaning of weight identification for each market, which is a component of composite information, is important for portfolio managers to help manage portfolios in response to various types of shocks.