The Usefulness of Non-earnings Financial Statements Items as Earnings Predictors

This Paper Provides empirical evidence on the predictive ability of nonearning annual report numbers under an earning prediction approach. The most studies report that earning were generated by a random walk Process. The paper investigates information Content of accounting items. The Logit prediction models were estimated based on pooled data set of 71 firms over the period 1371-79. The result demonstrates the predictive power of accounting item to explain and Forecast of earnings. However the predictive Performance model in the period 1377-79 was not significantly different from random walk model; except in 1378. Gross margin ratio, growth in Assets per share, growth in Assets, growth in Sales to assets ratio, total debt to total assets, pretax income to sales, growth in net income to sales, growth in operating expense to sales and changes in return on equity, correlated with earnings and these explanatory variables whose estimated coefficients had a wald Statistic significantly different from zero at the 10% level. The finding suggest that a firm's nonearings annual report numbers contain information concerning the direction of its next year's earnings change.