| Summary: | 碩士 === 國立臺灣科技大學 === 財務金融研究所 === 105 === This research uses the Taiwan’s company as the sample, the sample’s period is 2000 to 2016. The research studies the relationship between earnings forecast quality and the reputation of financial analyst’s institutions, the trend of earnings forecast quality, the interaction between the reputation of financial analyst’s institutions and the volume of underwriting market, and verify the relative measures that to improve earnings forecast quality are effective or not.
The empirical results of this study show that the trend of weekly earnings forecast bias is declining as a whole, which is consistent with O'Brien's (1988) and Brown (1997) literature that the earnings forecast tend to be more optimistic in the long-term, while the short-term earnings forecast tend to be more pessimistic. The weekly earnings forecast accuracy trend is consistent with the arguments presented by the literature of Elton, Gruber, and Gultekin (1984), Calderon (1993), Lim (2001) and Barron, Byard, and Liang (2013) that are positive relationship between the gap that between actual earnings date and predict announcement date and the earnings forecast accuracy.
The empirical results of this study also show that the reputation of the financial analysis institution is inversely related to the earnings forecast bias; the reputation of the financial analysis institution is inversely related to the earnings forecast accuracy. And, the the relative measures that to improve earnings forecast quality are effective. Finally, it is found that the financial analysis institutions (securities firms) that having good reputation are more optimistic and inaccurate when the underwriting market is hot. When the market is hot, the financial analysis institutions (securities firms) that having not so good reputation its earnings forecast is also more optimistic and inaccurate.
|
|---|
