비정상 거래량의 측정 방법론에 관한 연구

In this paper, we introduce several methodologies for measuring abnormal trading volume and compare each methodology's characteristics and the measuring performance of abnormal trading volume. As methodologies that utilize market trading volume, we present the OLS estimation method and GLS estimation method based on regression analysis and a method for calculating the abnormal trading volume by subtracting the market trading volume from individual trading volume. For the methodologies that do not use market trading volume information, we introduce the method of subtracting the average trading volume of stock from the trading volume of the stock, the method of calculating the ratio between the trading volume of a stock and the average trading volume of the stock, and Gervais et al. (2001) methodology. In the empirical analysis, we measure the abnormal trading volume for the Korean stock market from 2001 to 2020 using each methodology and comparing the performance. From this analysis, we show that the rate of judging the abnormal trading volume of the same stock as an event of extremely high or low trading volume differently reached 5.9 to 66.4%, and there is a significant difference in the results of comparing the performance of portfolios made of these stocks. These results suggest that the abnormal trading volume measurement should be carefully selected according to the purpose of the analysis.