The Research on Business Income Tax Related Selectional Cases

• Main Authors: Chih-yueh Tsai, 蔡志岳
• Other Authors: Yuh-Jiuan Parng
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/41680944385213135720

碩士 === 亞洲大學 === 會計與資訊學系 === 103 === Abstract Income Tax and Sales Tax belong to different taxable items as well as tax regulations, nevertheless all the Income tax collection orBusiness Income Tax with its techniques and detailed issue mostly refers to Sales Tax related data, there are principles of commonality and duplication avoidance between the information of these two taxable items. In this study, per the agreement and cooperation of tax accounting dealers, about 600 samples of customers’ tax declaration are gathered to be surveyed. Firstly, Logistic Regression is used to draw out possible key factors. Then Data Mining Model is used to continuously find out a decision tree analyzing path in order to predict the likelihood of those selected and investigated cases. To discover the possible key factors to cause abnormal tax eclaration is theoriginal expectation of this report. Furthermore, to build an immediate early warning system to help customers find unreasonable problems and related issue ASAP, to reduce the risk of illegal tax reporting and to develop customer’s confidence to CPA’s office or Accounting CPB with professional image are the following purposes. The conclusions throughout this research tell which factors have visible influence on and which not. Among the factors of independent variables, It is concluded that The Stock, income value, sales mount, VAT rate, shareholders equity (credit) have visible influence. Among the factors of control variables of various industries, the trading industry is significant but the other not. Result of empirical research is relevant to meet the expected outcome. Key words: Sales Tax, VAT rate,