OFFSET method based on logistic regression

• Main Authors: HE HUIZHU, WEI HU
• Format: Patent
• Language: Chinese; English
• Published: 14.01.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses an OFFSET method based on logistic regression. The method comprises the following steps: step 1, confirming default factors influencing a client, including operator data, consumption behavior data, loan application information, terminal equipment information, active geographic position information, credit score and the like; step 2, collecting, converting, quantifying and deriving related data; step 3, constructing a model for iteration and operation; step 4, verifying the effect. According to the method, logistic regression and an offset method are combined to solve the problem of systematic feature data loss of part of data sources in a certain time period, and a default model is constructed step by step and instance verification is performed. From the perspectiveof result comparison, after the offset method is applied, the stability and accuracy of the model are obviously improved. 本发明公开了一种基于逻辑回归的OFFSET方法。本发明实现步骤如下：步骤1、确认影响客户的违约因素，包括运营商数据、消费行为数据、借贷申请信息、终端设备信息、活跃地理位置信息和信用评分等；步骤2、相关数据