Trading volume and accounting research: Theory, implication, and evidence
Abstract
The use of trading volume data for capital market research has been relatively limited, due mainly to the lack of a theory for trading volume. Despite the importance of volume data recognized by Beaver (1968) and other researchers, theoretical guidance for information content studies on trading volume is still lacking. The theoretical part of this study attempts to explain the trading volume reaction to an informational event by using a simple general equilibrium model. Specifically, this study shows that an (abnormal) volume reaction to an event is caused by heterogeneous changes of beliefs among investors, and that the magnitude of trading volume reaction is positively related to the dispersion rather than the magnitude of changes in beliefs among investors. Furthermore, it is argued that to investigate empirically the information content of an event, a price effect study alone is not always sufficient; a complementary volume effect study is necessary. In addition, with various hypothetical cases I illustrate how a price and a volume reaction should be interpreted in conjunction with each other. In the empirical part of this study, one of the theoretical results, i.e., the positive relationship between the magnitude of trading volume and the dispersion of investors' belief changes, is empirically tested. Focusing on the abnormal trading volume reactions to earnings announcements, two proxies for the dispersion of belief changes are identified: dispersion of earnings forecasts by financial analysts (DEF) and analysts' average earnings forecast error (EFE). With the proxies two empirical hypotheses are derived. Based on the 464 observations from 204 firms, the hypotheses are tested through regression analysis as well as parametric and non-parametric correlations. Other firm-specific characteristics which influence trading volume reactions at earnings announcements are also described and discussed. The results support the theory of association between dispersion of belief changes and trading volumes.
Degree
Ph.D.
Advisors
Ro, Purdue University.
Subject Area
Accounting
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.