指導教授：曾智揚臺灣大學：企業管理碩士專班柳蔚柔Laowiriyakul, HataitipHataitipLaowiriyakul2014-11-302018-06-292014-11-302018-06-292014http://ntur.lib.ntu.edu.tw//handle/246246/263745Earnings Management is an important and interesting topic since it helps undermining the credibility of financial statements. There are several academic researches that try to use different proxies to detect the firm that manipulate their earnings. From the selection of these papers, I decided to use the DuPont analysis as the diagnostic of earnings management. In order to compare and contrast with the prior research on U.S. observations, I will focus on the five East Asian countries which are China, Hong Kong, Japan, South Korea and Taiwan. From the above interesting motivation, I have set up 3 hypotheses in order to test the usefulness of this diagnostic. First, I hypothesizes that the firms that meet or beat their threshold are more likely to have managed earnings upward which causes an increase in profit margin and a decrease in asset turnover. Second, I will test whether the firms that miss or beat their threshold by large amount are more likely to have managed earnings downward. Third, I predict that if a firm manipulates earnings upward, future profitability should fall as the income contribution from earnings management in the current period reverses. According to the findings, some results are consistent with my expectation especially my first hypothesis and when separate the observation into the country basis which significantly prove that the firms that meet or beat their threshold are more likely to have managed earnings upward. When separate analyses into the country basis, it can better support my main hypotheses, and at the same time, provides more expectable reasoning on the rejected ones.Chapter 1 Introduction 1 Chapter 2 Literature Review and Hypothesis Development 2.1 Measurement of Earnings Management in Literature 4 2.2 Hypotheses 8 2.3 Limitation of the DuPont Analysis in Detecting EM 9 Chapter 3 Empirical Design 3.1 Sample and Data 11 3.2 Variable Definition 12 3.3 Descriptive Statistics 16 3.4 Empirical Models 22 Chapter 4 Results 4.1 Univariate Analyses 23 4.2 Multivariate Analyses 25 Chapter 5 Additional Analyses 30 Chapter 6 Summary and Conclusions 36 References 38742607 bytesapplication/pdf論文公開時間：2014/07/22論文使用權限：同意有償授權(權利金給回饋學校)使用杜邦分析偵測盈餘管理thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/263745/1/ntu-103-R01749056-1.pdf