A Generalization of the API Methodology As a Way of Measuring the Association between Income and Stock Prices
针对会计研究中广泛使用的异常绩效指数（API）的缺陷，提出一种推广方法，以更有效地衡量收入与股票价格之间的关联，对会计和金融领域的实证研究者有参考价值。
For over a decade now, accounting researchers have employed the Capital Asset Pricing Model (CAPM) to evaluate empirically the association between market prices and accounting information. Most studies have used some variant of the Abnormal Performance Index (API),1 introduced into accounting literature by Ball and Brown [1968]. Although the API has been widely used, several of its deficiencies have been recognized, for example: (a) What constitutes good or bad is largely arbitrary. (b) Only the sign of the news has been used but not its magnitude. In a recent study, Beaver et al. [1979] examined the association between unsystematic security returns and the magnitude of earnings forecast errors. This association was found to be statistically significant, thus pointing out the potential usefulness of the magnitude of the news. (c) Usually, only firm-specific news has been used, with the interrelationship among firms being neglected. (d) A well-defined