Data of Taiwanese: Firms Shifting Earnings from Small Losses to Small Profits
Description
We share the original files used in the paper, and each XLSX files is the original file needed to create diagrams and tables.
Files
Steps to reproduce
We will Individually describe the data files and how to use them to build tables and graphs. 1. Figure1 presents the number of companies in different EP ratio groups. We collect the Earnings-Price ratio of each company at the end of the month preceding the earnings announcement and group by 0.0125. 2. Table1 reports the number of observations and firm characteristics in 2005-2018. 3. Table2 reports median earnings management (EM) measures of Taiwanese firms from 2005 to 2018. EM is estimated by discretionary current accrual (DCA) using Cohen and Zarowin's (2008) method. 4. Table3 reports cumulative abnormal returns (CAR, raw returns minus required returns) from day 1 to day 1 relative to the earnings announcement, in which required returns are calculated by the market model using data from day 65 to day 6. There are 577 and 1,846 observations for the small-loss and small-profit groups, respectively. We use two criteria to select firms that possibly engage in loss-to-profit EM, resulting in 517 and 214 observations. 5. Table4 uses three sets of samples to conduct regressions with the yearly effect: (1) both the small-loss and small-profit, (2) the small-loss (577), and (3) the small-profit (1,846) firms. The output variable is cumulative abnormal returns (CAR(1,1)) around earnings announcements. 6. Table5 reports the factor loadings from a factor analysis used to reduce the number of variables that measure information symmetry and corporate governance. The sample contains 15,207 observations from 2005 to 2018. 7. Table6 reports the correlation matrix of 517 loss-to-profit and 577 small-loss observations from 2005 to 2018. 8. Table7 reports the results of the logistic regressions using a dummy equal to 1 if firms manage earnings from small losses to small profits, and 0 if firms have small losses. There are 577 small-loss and 517 loss-to-profit observations from 2005 to 2018.