Return to search

我國上市公司獲利能力指標與股價變動之研究

本研究係以各種獲利能力指標,衡量企業在一特定期間內之獲利能力情形與股價變動的關係,企圖以獲利能力指標來解釋股價變動,以充分瞭解投資者如何使用獲利能力指標等相關資訊。本研究之研究包括下列幾點:首先,可用以衡量企業在一特定期間獲利能力的指標有那些,又,這些指標彼此間的關係為何?第二,本研究所選取之獲利能力指標,是否對股價變動有充分的解釋能力?第三,會計盈餘是企業於一特定期問的獲利情形,但過去研究指出:以未預期盈餘解釋股價變動,解釋力往往偏低,同時亦無法充分衡量企業績效(Fisher and McGowan, 1983);而獲利能力指標既然表彰著一企業於特定期間的營運結果及其他績效結果,則當以未預期盈餘以外之獲利能力指標來解釋股價變動時,是否要比未預期盈餘對股價的變動,具有更高的解釋能力?第四,應用盈餘反應係數模式,即在獲利能力指標/股價關係模式中,若加入Collins and Kothari(1989)的獲利能力持續性、經濟成長機會與系統風險等因素後,該獲利能力指標是否更能解釋股價的變動?
由於我國股市中,績優股與投機股的股價行為迥異,為得到較為一般化的結果,本研究乃按CAR的值分為三組,分別是CAR為最大的一組(符號為正,樣本童為285)、CAR為最小的一組(符號為負,樣本童為280),以及介於上述兩組之間者(樣本量為280)。這種樣本的區分方式,對於CAR絕對值較大的二組,本研究定義為投機股;CAR值居中的一組,則定義為績優股。本研究對於所定義的投機股,又分別按其正、負符號分成兩組,蓋因CAR絕對值較大的樣本,其股價行為對於資訊的敏感程度較高,多、空資訊的發佈(包括會計盈餘及非會計盈餘)容易造成其股價變動,且多、空之股價行為不同;本研究以為這二組樣本特性應有很大的歧異性,故特別將CAR絕對值較大的樣本分為正、負二組,主要在區別這二種情形。惟就實際情形而言,CAR絕對值較大者不見得就是投機股;CAR值居中者亦不見得就是績優牛皮股。此處僅就直覺分類,實際上是否如此,仍有待進一步研究。
本研究所獲致之結果如下:
一、在各獲利能力指標彼此間之關聯性方面,本研究就13個獲利能力指標做相關分析的結果,大致上支持這些獲利能力指標間存在顯著正相關。
二、就各獲利能力指標與股價變動之關聯性而言,除CAR為居中的一組較難獲致結論外,CAR為最大的一組及CAR為最小的一組,大致上皆證明獲利能力指標對股價變動有重大影響,但解釋能力不高。此外,本研究曾企圖將獲利能力指標區分為經常性及非經常性兩種,推論以經常性盈餘為計算基礎者,應能對股價有更佳的解釋能力,然而就實證結果觀之,並未支持此項推論。
三、就各獲利能力指標相對於未預期盈餘對股價變動之資訊內涵而言,除CAR為居中的一組並未獲致具體的結論外,其餘兩組大致上均能支持獲利能力指標變數確實比未預期盈餘變數更能夠解釋股價的變動。
四、若將各獲利能力相標變數,取代盈餘反應係數模式中之預期盈餘變數,再加上其他控制變數如系統風險、經濟成長機會及盈餘(獲利能力)持續性等,經由實證結果發現,除CAR為居中的一組難以獲致具體的結論外,其他各組均證實以獲利能力指標變數取代未預期盈餘變數後,其對股價變動解釋能力,要比單就未預期盈餘本身或單就各獲利能力指標本身有相對較高的解釋能力。 / The purpose of this study aims to investigate how investors apply profitability measures in their investment decisions. For example, does an investor prefer accounting rate of return or returns on investment or other specific profitability measure to evaluate the change in firm's value? This is the main research question of this study. Chen (1992), Chiau (1995) and others demonstrated that accounting earnings is an useful information but its usefulness is quite unstable. Therefore, a profitability measure other than accounting earnings may be the one causing the unstable results. In order to obtain a generalized empirical result, this study compares classic (for example, Ball and Brown, 1968) to ERC (Earnings Response Coefficient) models (Collins and Kothari, 1989). The results of this study can be used for explaining which profitability measure is more useful in Taiwan stock market. In addition, the results can provide a relative explanation whether ERC model is an appropriate one or not.
The samples elected from listed companies in Taiwan stock market between 1991 to 1995 during which Taiwan stock market enjoyed a stable market environment. This character fits the requirements of hypotheses. In addition, in order to properly describe price characteristics, this study applies a statistical classificatory method to classic all samples into three groups underlying the properties of CAR (Cumulative Abnormal Returns).
The empirical findings can be summarized as follows.
.Each defined profitability measure, including accounting earnings, can significantly explain the variability of stock returns. In addition, there exists significant correlation among defined profitability measures.
.The classification of samples underlying CAR does increase the explanatory power of profitability/return relationship, hi particular, the larger and smaller CAR samples exhibit better profitability/retum relationship than medium group does.
.No matter whether classic or ERC model is applied, the empirical results are similar each other. Thus, the ERC model may not be a generalized model in Taiwan stock market. It requires more research to develop a generalized empirical model for substituting ERC model referring the special characters of Taiwan stock market. The theory of economics of informational asymmetry may be the one.

Identiferoai:union.ndltd.org:CHENGCHI/B2002002467
Creators楊炎杰
Publisher國立政治大學
Source SetsNational Chengchi University Libraries
Language中文
Detected LanguageEnglish
Typetext
RightsCopyright © nccu library on behalf of the copyright holders

Page generated in 0.0021 seconds