Myers(1999)實證研究發現,Ohlson (1995)與 Feltham and Ohlson (1995,1996)所發展的會計基礎評價模式(包括線性資訊模式與評價模式)在美國股市未獲得支持。本研究旨在探討此會計基礎評價模式是否適用於臺灣股市。由於線性資訊模式與評價模式之係數與公司所處的經濟背景、生產技術及會計政策等公司特性有關,因此本研究針對個別公司使用時間序列方式估計迴歸係數,並以民國80年至90年臺灣股票上市公司季資料為樣本,實證結果發現: 1.就線性資訊模式而言,Myers(1999)研究顯示,線性資訊模式所估計的係數僅部份獲得支持;本研究則顯示,線性資訊模式所估計的係數則全數獲得支持。因此,線性資訊模式在臺灣股市適用情況優於美國。 2.Myers(1999)研究結果顯示,藉由線性資訊模式所估計之公司價值(稱為估計價值)遠低於公司的股價,其可能原因為截距項是負的,以及市場有捕捉到未來預期的異常盈餘,而會計資訊則未捕捉到。評價模式估計係數與線性資訊模式所設算之係數並不相等,顯示會計基礎評價模式並不符合內部一致性。 3.就評價模式而言,本研究以非財務指標與市場競爭結構變數當作其他資訊,非財務指標與市場競爭結構變數加入評價模式後,顯著提高對股價的解釋能力。 4.就保守會計係數而言,依據線性資訊模式所估計的保守係數確實能夠衡量公司保守會計的程度。
Myers(1999) found that the residual income valuation model developed by Ohlson(1995), Feltham and Ohlson (1995,1996) was not verified in the stock market of the United States. This study intends to explore if the residual income valuation model can be applied to the stock market of Taiwan. This study estimates the parameters of the LIMs as a time series rather than in the cross-section because the parameters of the LIMs must be a function of the firms’ economic pressures, production technology and accounting policies. We have taken the listed companies in Taiwan from the first quarter of 1991 to the fourth quarter of 2001 as examples. The research results show as follows: 1.As to the linear information dynamics, Myers(1999) indicated that only part of the parameters estimated in LIM were supported. However, our study indicates that all parameters in LIM are supported oppositely. Hence, the limitations of LIM are more adequately useful in Taiwan’s stock market than that in the US. 2.Myers(1999) indicated that the firm value estimated by linear information dynamics is significantly less than the stock price. Such an understatement of value occurs because the negative intercept, and because the market fully captures the expectations of future RI. The models fail to capture the market’s expectations of future RI. The coefficients on the information variables in the price level regressions are not equivalent to the median valuation coefficients implied by the estimated coefficients in the linear information models. 3.As to the valuation model, this study regards non-financial indicators and competitive market structure variables as the other information of Ohlson model, which will significantly enhance the explaining capability towards the stock price. 4.As to the Conservatism coefficient, the Conservatism parameters estimated by LIM are fully capable of measuring the degree of conservativeness of a company’s accounting principle.