The Association between Financial Statements Text Analytics and Financial Warning

• Main Authors: Lin, Kai-Ming, 林鍇銘
• Other Authors: Seng, Jia-Lang
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/psydye

碩士 === 國立政治大學 === 會計學系 === 107 === When analyzing financial statements for investment, investors are most worried about invest potentially dangerous companies. Company financial crisis or the financial fraud occurred after investors holding shares to damage the price. Therefore, for investors, immediate financial warning is very important. The purpose of this study is to investigate whether there is a significant different for company in distress and non- distress in management discussion and analysis in annual reports. This study chooses company which have being change transaction method by Taiwan Stock Exchange or have committed frauds from January 2016 to March 2019. Manually interpret will be used to analyze Management Discussion and Analysis in annual report and calculate relevant sentiment tone words. Finally, developing models to predict potential risks for companies. The empirical results show that company in distress has lower proportion of positive words, higher proportion of negative words, lower proportion of emotional words, lower proportion of punctuation, and less examples for economic conditions in the annual report.