Gassama Abdoul Gadiri
University of Science and Technology of China A dissertation for master degree Ordinal Probit Regression Model and Application in Credit Rating for Users of Credit Card Author : Speciality : Supervisor : Gassama Abdoul Gadiri Probability and Statistics Prof. HU Taizhong Finished Time : April, 2012
( Credit Card) Probit, Probit I
ABSTRACT ABSTRACT Credit is the inevitable product of social development and we can say there is no credit system, there is no current economic boom Development. With the development and growth of the credit system, more and more institutions and individuals is included into the credit body. The demand for credit ratings is a top priority. Faced with a limited number of individuals and institutions, individuals will have enough energy to know the credit of these institutions and individuals and then make decisions. By Economy is booming, the credit system tentacles deep into almost every aspect of everybody s life, thus there is no effort to get them to understand whether individuals or institutions will have a good credit record. Credit Rating often uses quantitative means to analysis credit history and gives individual or institution a credible Rating. Credit Card is a common credit products, also is a kind of credit services, distributed by bank or credit card company in accordance with the user s credit rating and financial resources and cardholders can overdraw from the card instead pay cash, and payback to the bank at the end of each term. The most important problem that credit card banks faces is about the default of cardholders. In this paper, ordered probit regression models are used to model the credit rating with a specific data sets, and the main factors for default of cardholders are found. Keywords: Credit Card, Credit Rating, Ordered Probit Model. III
1 1 1.1 1 1.2 2 2 5 2.1 5 2.2 5 3 Probit 9 3.1 Probit 9 3.2 11 3.3 12 3.4 14 17 19 V
1 1 Probit Probit 1 2 3 Probit 1.1 (Credit Rating) 20 1902 1
1 1.2 2011 8129.56 110.31 4% [1][2] ( KMV ) 2
1 3
2 2 ([3]) www.rdjg. com.cn 2.1 8372 R head?? 7390 0.08 10000 2680 0 0 0 6036 1.00 5000 0 1 0 0 807 0.00 2000 0 1 0 2 5154 0.95 5000 0 5 11 1 3616 0.36 3000 851 0 4 3 5140 1.01 5000 0 8 5 4 8 2.1 2.2 2.2 ( ) ( ) X 1 = 5
2 Y 0 1 1 30 2 31 60 3 61 90 4 91 120 5 121 150 6 151 180 7 180 2.1. 2.2. 2.3 X 2 X 6 X 2 X 6 Y = y 2.4 2.5 2.4 Y ( ) X 2 ( ) X 5 ( ) X 6 ( ) Y ( ) X 3 ( ) X 4 ( ) 2.5 6