The association between trading volumes and analyst’ belief structure around earnings announcements

• Main Authors: Li, chi hung, 李啟弘
• Other Authors: Chang, Bao Guang
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/07660649567294561545

碩士 === 淡江大學 === 會計學系 === 92 === Abstract: Beaver (1968) indicate that price changes reflect the change in the aggregate market’s average beliefs, while trading volume is the sum of all individual investors’ trades. Given the interpretation, we can understand there is enormous difference in concept. Bamber and Cheon (1995) found nearly a quarter of the earnings announcements generate either (1) very high trading but little price change, or (2) large price change but little trading. The study tells us that if we control for the price change, trading volume may express additional information. Research scholars believe the addition information come from investors’ belief revisions. As for how do researchers capture investors’ belief revisions, Bamber, Barron and Stober(1997) find three distinctly different aspects of disagreement each play an incremental role in explaining trading volume, even after controlling for the magnitude of the contemporaneous price change. These aspects of disagreement are: dispersion in prior beliefs, change in dispersion, and belief jumbling. Dispersion in prior beliefs is the level of variation in expectations before the earnings announcement, change in dispersion is the difference in the level of dispersion in beliefs after vs. before the earnings announcements, and belief jumbling occurs when investors’ beliefs change position relative to each other around earnings announcements. And our study use the variable called “consensus”, which is introduced in Barron, Kim, Lim, and Stevens (1998) (hence BKLS model), to further capture the concept of investors’ belief revisions. BKLS model think each analyst’s available information about earnings consists of common (public) and idiosyncratic (private) information. When analysts rely more on their idiosyncratic information to create earnings forecasts, the so-called consensus value will decrease, and vice versa. The concept of consensus value is the degree to which analysts rely on common information. So the consensus value is relative, not absolute.The result of our study is that the consensus value is significantly related to trading volume after controlling for other volume-related effect. We hope that the result will give some contribution to researchers when they try to explain trading volume.