투자자별 거래정보가 주식수익률에 미치는 영향

The purpose of this paper is to verify whether the net buying factor of each investor can be evaluated as a risk factor that can explain the common variation of stock returns. For the basic analysis, cross-sectional correlation analysis and regression analysis between characteristic variables and net buying ratio of each investor were conducted to examine the differences in trading behavior by subject. We also examined the characteristics of portfolios based on the net buying ratio of institutions, individuals and foreigners. Based on this basic analysis, three factor mimicking portfolios were constructed: TLMH, DLMH, and FLMH. And We verified that these factors represent a systematic relationship with the testing portfolio. To verify whether the net buying factor of each investor has a risk premium, a testing portfolio was constructed based on the coefficient for the net buying factor of each investor, and then the risk-adjusted return was measured to further analyze whether there was a difference in returns. The main results of this paper are as follows. First, as a result of time series analysis of the FF3 and FF5 factor model by adding the net buying factors of each investor, the factor coefficient of the net buying factors of each investor showed a systematic change according to the level of the net buying ratio of each investor. Second, most of the intercepts of the testing portfolio were statistically insignificant or near zero. Third, the explanatory power of the model remained high. As a result, we can confirm the empirical results that can be judged as a risk factor that can capture the change in stock returns.