Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Rejection inference method based on Cox regression and logistic regression and electronic equipment

A logistic regression and algorithm technology, applied in the field of financial science, can solve the problems of parameter estimation, inability to distinguish whether bad samples are overdue in the current period, and human subjective factors.

Inactive Publication Date: 2020-10-20
睿智合创(北京)科技有限公司
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

(3) Non-random missing, which depends on the incomplete variable itself, can be divided into truncated missing, censored missing, sample selection missing
[0012] 3. The user's overdue status during the observation window period is not used, and the binary classification variable cannot distinguish whether the bad sample is in the current overdue status during the observation window period for bad samples
Since the distribution is judged by the graphical method, there are human subjective factors, which will have a certain impact on the estimation of the parameters

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Rejection inference method based on Cox regression and logistic regression and electronic equipment
  • Rejection inference method based on Cox regression and logistic regression and electronic equipment
  • Rejection inference method based on Cox regression and logistic regression and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] This embodiment discloses as figure 1 A rejection inference method based on Cox regression and logistic regression is shown, including the following steps:

[0053] S1 collects all application user data within a preset period, and defines two sets of tags for each credit user, namely, the binary classification target variable and the survival analysis target variable;

[0054] S2 performs Cox regression modeling on the useful user data based on variables defined by survival analysis;

[0055] S3 calculates the probability P(G|A) of a "good" sample and the probability P(B|A) of a "bad" sample of a rejected sample based on the Cox regression results;

[0056] S4 Based on the inference results of binary classification labels and rejected samples, use the binary classification algorithm to train the model and complete the development of the scorecard model.

[0057] In S1, the observation window is set as MOB6, and each user has two sets of target variables. In the binary...

Embodiment 2

[0083] In this embodiment, the main steps and the derivation of related formulas are as follows,

[0084] 1. In the data preparation stage, collect all application user data within a preset period, and formulate target variable definitions for the vintage analysis and rollover rate analysis of useful credit customers, combined with the business scale of existing users after lending. Under normal circumstances, the observation window is set as MOB6 (the user uses the credit card as the starting point of observation for the first time, and the end point of observation at the end of 6 months).

[0085] Each user has two sets of target variables. In the binary classification algorithm, the definition logic for the customer group with useful letters is as follows. Users with IOUs that have been overdue for 30 days or more in MOB6 are defined as bad samples, assigned a value of 1, and there are loans in MOB6 And the user who has never overdue is defined as a good user, assigned a va...

Embodiment 3

[0101] This embodiment discloses an electronic device, which includes a processor and a memory storing execution instructions. When the processor executes the execution instructions stored in the memory, the processor executes rejection based on Cox regression and logistic regression Inference method.

[0102] In summary, the present invention uses the survival analysis model to infer rejected samples through the sample inference technology. First, it contains all sample data as much as possible. When the binary classification algorithm trains the model, gray samples are generally eliminated, and the survival analysis can be very fast. It is easy to handle samples with different overdue days and the performance period is not long enough. Secondly, the target variable contains more information. The survival analysis model directly fits the original features and censored status that define good or bad, that is, the overdue days and overdue status. Compared with the two In terms ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to the technical field of financial science, in particular to a rejection inference method based on Cox regression and logistic regression and electronic equipment. The method comprises the following steps of: S1, collecting all application user data in a preset period, and defining two sets of tags for each information user, i.e., a binary classification target variable anda survival analysis target variable; S2, performing Cox regression modeling on useful information user data based on variables defined by survival analysis; S3, based on a Cox regression result, respectively calculating a probability P(G|A) that a rejected sample is a 'good' sample and a probability P(B|A) that the rejected sample is a 'bad' sample after credit granting and loan passing; and S4, based on the binary classification label and the rejected sample inference result, training a model by using a binary classification algorithm to complete the development of a score card model. According to the method, the survival analysis model is used for deducing a rejected sample, a non-parametric method Cox regression analysis is selected, a distribution function of the survival duration doesnot need to be judged, the process is simplified, deviation caused by unreasonable selection of the distribution function is avoided, and it is guaranteed that the model effect is more accurate.

Description

technical field [0001] The invention relates to the field of financial science and technology, in particular to a rejection inference method and electronic equipment based on Cox regression and logistic regression. Background technique [0002] With the development of technologies such as big data analysis, artificial intelligence, Internet of Things and blockchain, the application of financial technology is subverting the development model of traditional industries such as finance, and realizing the transformation from offline to online Internet finance. In recent years, the consumer finance industry has a large number of traffic opportunities on the Internet to obtain customers, and a large amount of traffic and benefits can be obtained in a short period of time by enclosing land. However, as the national financial industry regulation becomes stricter, a series of special The introduction of rectification activities and regulatory policies has accelerated the reshuffle and...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06Q40/02G06K9/62G06F17/18G06F17/15
CPCG06F17/18G06F17/15G06Q40/03G06F18/24
Inventor 黄建王云清庄泽铭
Owner 睿智合创(北京)科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com