The Acccuracy Of Managements’ Earnings Forecasts

Wen Bach said:"In the world which has rapid change, the value of predictive information is more important than before."1990s, ICAEW, ICAS, AICPA have been raised about the research results of the future of financial reporting model, and the forecasts are said to be of great importance, the future financial reports should include the company’s forward-looking information and the information which could help investors and other users to predict and evaluate the company’s future financial condition. As an important part of the predicted financial information, the importance of Earnings Forecasts is obvious.With the continuous improvement of China’s capital market, investment in order to find quality companies from a number of listed companies. They are not satisfied with the historical financial information as before. They demand for forecast information, especially for earnings forecasts. Earnings forecasts are also of great significance for the capital markets and the enterprises themselves. For capital markets, earnings forecasts are informative. Disclosure of earnings forecasts conducive to the fair of capital markets and improve the effectiveness of the capital market. For enterprises, the earnings forecast information can be passed to the user as a signal of managements’conditions.Earnings forecasts refer to certain assumptions and premises, a reasonable forecast of future corporate profitability. The information is uncertainty and subjectivity, but the information’s correlation is higher. Listed companies, investors and capital markets need earnings forecast strongly. In order to meet the information needs of their investment decisions, the information users need to know the reliability of the information. If managements do not have a high reliability, so investors will suffer losses. Thus it is necessary to discuss the reliability of the information which is disclosed by management of listed companies. This article will substitute the performance forecast for the earnings forecasts of the management of listed companies. This article will study the accuracy of the managements’earnings forecasts in recent years by the establishment of absolute prediction error model, relative prediction error model, and mean absolute prediction error model. Through the establishment of multiple regressions, stepwise regression model to study the impact factors of the managements’ earnings forecasts.This article is divided into six chapters, the main content and framework as follows:The first chapter is introduction. Through the introduction of relevant background and significance of the earnings forecast to reflect the importance of reliability management earnings forecasts and research value. Then the research models and methods are generally introduced. Followed is a diagram of research ideas, arrange of the contents. The last is the major contribution of this paper.The second chapter is a review of the literature. Literature, from China and other countries, about the accuracy of management’s earnings forecasts and its impact factors are reviewed and compared. It is found that although the content of the study of managements’earnings forecasts is expanding and deepening, but the studies which substitute the performance forecast for the earnings forecasts of the management of listed companies is still less and few studies recognize that part of the performance forecast is in line with earnings forecasts.The third chapter is the institutional background and theoretical basis. In this chapter the author explain why substitute the performance forecast for the earnings forecasts of the management of listed companies and the importance of management’s earnings forecast.The fourth chapter is the study design. It comes up with hypotheses bases on the company characteristics, the financial characteristics, the profitability characteristics, the time characteristics, and the two control variables. Then these factors were explanations and their formulas were listed. Then establish the direct prediction error model, the absolute prediction error model, the mean absolute prediction error model, and establish a multiple regression model based on the absolute prediction error model. Finally, the samples and data were described.The fifth chapter is the process and the results of the empirical analysis. This chapter is divided into two parts. The first part is a descriptive analysis. The results indicates that:(1) the accuracy of the managements’earnings forecasts in recent years need improve, there is a certain distance between the higher standard;(2) management with "sound " feature;(3) China’s listed companies with " glossy " tendencies;(4) industry is the impact factors of the managements’earnings forecasts. The second part is the regression analysis:(1) The results said the current ratio, financial leverage, squared of sales growth and earnings per share is positive correlation with earnings forecast reliability;(2) Company listed term, forecast period is negatively correlation forecast earnings reliability;(3) The results about company size are not consistent in the overall regression and stepwise regression;(4) the control variable of annual has a significant impact on the reliability of predicted.The last Chapter is conclusions, recommendations and thinking. This chapter is divided into three parts. In the first part, the results of this analysis are summarized. The second part puts forward some policy suggestions on the basis of the conclusions. The third part points out the lack of this article and presents some ideas for future research.The main contributions of this paper are:first, the study has some innovative angles. This article substitute the performance forecast for the managements’ post-IPO earnings forecasts and take advantage of descriptive statistical analysis and regression analysis of listed companies management earnings forecast reliability status and influencing factors were studied. Second, the research methods have some improvements. In the study, the independent variables which have little effect on the reliability of predicted were excluded, and in the study Join the control variables of annual to represent external factors such as macroeconomic factors and institutional change factors.