Please use this identifier to cite or link to this item:
Multiple Impacts of an Earnings Announcement on Security Returns
Chang, Conrad C.
Earnings announcement;Earnings level;Earnings growth;Dynamic linear growth model;Earnings response coefficient
|Issue Date:||2017-11-15 14:54:06 (UTC+8)|
|Abstract:||大部份有關資本市場的會計研究都藉著非預期報酬對非預期盈餘的簡單迴歸分析方法，來探討會計盈餘的資訊內涵。這個簡單迴歸係數稱之為盈餘反應係數(Earning s ResponseCoefficient, ERC), ERC能夠衡量股價對會計盈餘宣告的反應敏感度。雖然大部份的ERC研究都已經探討了未來盈餘水準之非預期增加對股價的影響，這些研究卻忽視了非預期盈餘對未來盈餘成長的影響。根據Gordon的股利成長模型，當盈餘宣告時，投資人不僅會調整對未來盈餘水準的預期，同時也會調整對未來盈餘成長的預期。為了彌補以往會計研究忽略盈餘成長之情況，因此本研究同時探討盈餘水準和盈餘成長的預期改變對股價的影響。本研究使用動態線性成長模型(Dynamic Linear Growth Model, DLGM)來衡量動態的盈餘水準和盈餘成長。研究結果發現盈餘水準更新和盈餘成長更新對股價均有顯著且幅度不儘相同的影響。亦即從一個公司的盈餘宣告可以得到兩個ERCs (ERC_(level)和ERC_(growth))，而且ERC_(level)小於ERC_(growth)。另外也發現，將非預期盈餘分解可以提高盈餘對股價的解釋能力。|
Most capital market studies in accounting examine the information content of accounting earnings by regressing unexpected returns on unexpected earnings. The slope coefficient of this simple regression is called the earnings response coefficient (ERe). The ERC measures the response sensitivity of stock prices to accounting earnings announcements. Although most ERC studies have examined the impact of the unexpected increase in future earnings levels on security prices, they overlook the influence of unexpected earnings on future earnings growth. According to Gordon's dividend growth model, when earnings are announced, investors should not only revise their expectations for future earnings levels, but also their expectations for future earnings growth. This study examines the impact of revisions of future earnings growth, in addition to those of future earnings levels, on security prices . This study applies the Dynamic Linear Growth Model (DLGM) to derive time-varying measures for both earnings levels and earnings growth. The DLGM also decomposes unexpected earnings into three earnings components: revisions in earnings levels, revisions in earnings growth and noise in earnings. This study finds that both revisions have significant but different impacts on security returns. Specifically, two ERCs (ERC_(level) and ERC_(growth) are derived and ERC_(growth) is found to be greater than ERC_(level) as theory predicted. Moreover, it is found consistent with the theoretical model; the empirical evidence shows the decomposition of unexpected earnings into earnings components increases the explanatory power of earnings on security returns.
|Relation:||會計評論, 29, 135-174|
|Appears in Collections:||[會計評論] 期刊論文|
Files in This Item:
All items in 學術集成 are protected by copyright, with all rights reserved.