A Study of EPS Forecasting Model and Financial Ratios

• Main Authors: Wu, Shin Yi, 吳信億
• Other Authors: Zhou, De Guang
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/54174528885423164269

碩士 === 南台科技大學 === 會計資訊系 === 92 === Financial statements are not only the result of the business performance, but also an important indicator of future perspectives, especially about how to use public financial statement information to predict the business profits, which is an important issue for both academic and practical purposes for long. However, how to secure or strengthen the “predictable value” from the financial statement information is not just a financial accounting professional ethics problem. The new variety of business models and transactions change so fast, and industrial environmental dynamic and complexity increase so rapidly, therefore imposing that providers and users must reform and adjust to such rapidly changing situation. Similarly, the process, regulation, form and context of the financial statements are often required to keep changing at the same pace keep up with the time. Is the information under the present structure still prediction? And how to use this data to sift the predictable information? These are all continuous and ongoing research issues to review and adjust. The purpose of this study is to identify the financial ratios that are predictors of performance and to compose EPS forecasting models with those financial ratios. Notice there are possible impacts of time span embedded in the forecasting models, as well as the selected predictors, we propose seven time span modes to build up the forecasting models and test its effectiveness separately. Based on our analysis of Taiwan steel and computer makers, we find that there are different sets of EPS predictors between the two industries. And as we expected, the effectiveness of the forecasting models with each time span mode are quite different. This could be an interesting direction for future research.