시장정보가 과연 일중 주가 변동성과 거래량에 반영되는가?

Abstract

본 연구는 하루 동안 올라오는 뉴스의 헤드라인 개수를 시장정보의 변수로 사용하여 정보와 일중 주가 변동성, 거래량과의 관계를 연구하였다. 일중 거래시간을 30분간으로 나누어 도착하는 시간대별 뉴스의 흐름 및 변동성과 거래량의 패턴을 규명하였다. 거래시간동안의 뉴스량의 패턴은 급격한 V자형의 패턴을 보였고, 변동성 및 거래량은 U자형에 유사한 패턴을 보여주었다. 공통적으로 오전 시간에 가장 높았고, 시간에 따라 감소하다 폐장 시간 때에 약간의 증가하는 모습을 보여주었다. 그리고 시간대별로 올라오는 뉴스의 양과 시간대별 주가 변동성과 거래량에 대해 회귀분석을 실행한 결과 주가 변동성 및 거래량이 뉴스의 양과 유의적인 시간대가 존재했다. 하지만 변동성보다는 거래량과의 유의적인 시간대가 좀 더 많이 나타났다. 추가적으로 거래량의 시간대를 시계열로 나열한 후, 개장효과와 폐장효과를 통제하고 회귀분석을 실시한 결과, 모두 1% 수준에서 유의적인 양의 관계를 보였다. 본 연구 결과는 국내 시장의 경우 직접적으로 정보 변수를 사용하여 일중 주식시장의 움직임, 특히 거래량을 설명할 수 있음을 보여주었다

Tliis paper explores the relationship between public news and the intraday market price volatility, trading volume in the Korean stock market. Financial economists have often found that the real impact of public news on the market volatility and the trading volume is not as big as apparently expected in the investors' side. The issue is related to the over- or under-reaction to the news in the capital market. We believe this study is the first one of its kind which deals directly with the news variable in the Korean stock market. The number of headline news in every 30 minutes during 2001-2003 from the E-daily is used as a proxy for the public news. The corresponding every 30 minute price volatility and trading volume are also matched to the public news variable. The apparent pattern of public news turns out to be a very sharp V-shape, while the volatility and the trading volume show a near U-shape pattern. A heavy volume of news arrives in the beginning of the market, and the magnitude of news decreases in the trading time and increases back in the closing time. In the US market, a reverse J-shape is observed which means the majority of market news comes to the market in the near closing time. The regression analysis using the 30 minute news arrival and the market price volatility, the trading volume has been performed. The major results of the paper are as following. First, the public news can explain the price volatility as well as the trading volume in the several time periods. It is interesting that, in the case of the US market, the news doesn't play any significant role in explaining the price vola仕lity and the volume as reported in Mitchell and Mulherin(1994), and Berry and Howe(1994).Second, the regression analysis shows that the public news plays some more meaningful roles in the behavior of trading volume rather than in the price volatility. The total number of significant time period for the trading volume and the news is 8, while the market price volatility and the news shows some significance in the 5 time period. This result is similar to the previous researches in which there is little significant relationship between the public news and the market activities especially in the price volatility. Finally, the paper concludes that the public news can be closely related to the Korean stock market behavior, especially in the trading volume. The result from the Korean market implies that public news in general can explain the market activities. In most of the previous studies in the U.S. market, news turns out to be weakly related to the market trading activity. And the price volatility is not closely related to public news in the U.S. market. Interestingly enough, the Korean market seems to show rather significant relation between the news and the market trading volume and the volatility. The limitation of the paper is the potential bias problem generated from the exogeniety of news variables in the data. Some news variables may directly describe the market activities and some other news variables may affect the market investment behavior differently in terms of direction as well as magnitude. Further meaningful studies using news variables separately in each category according to the contents of the news has been suggested.