CN110533520A - A kind of ranking method of the individual customer overdue loan grade based on multi-model - Google Patents
A kind of ranking method of the individual customer overdue loan grade based on multi-model Download PDFInfo
- Publication number
- CN110533520A CN110533520A CN201910490288.8A CN201910490288A CN110533520A CN 110533520 A CN110533520 A CN 110533520A CN 201910490288 A CN201910490288 A CN 201910490288A CN 110533520 A CN110533520 A CN 110533520A
- Authority
- CN
- China
- Prior art keywords
- overdue
- risk
- grade
- grades
- client
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The ranking method for the individual customer credit risk grade based on multi-model that the present invention relates to a kind of, the following steps are included: in client's overdue 7 days, 14 days, 21 days and 30 days periods, overdue feature of the client in each overdue stage is obtained, respectively building violation correction model corresponding with each overdue stage；Violation correction model based on above-mentioned building establishes overdue risk class model stage by stage；The overdue feature of current overdue client is obtained, and according to the client in different overdue stages, calls corresponding violation correction model；Construct the risk of loss grade of current overdue client；Based on default risk grade and risk of loss grade, the composite rating of overdue client is obtained.The present invention provides a kind of ranking method of individual customer credit risk grade based on multi-model, it can be handled for overdue client's differentiation, preferentially the grading higher client of risk is handled, manpower is saved, promotes collection efficiency and customer experience, prepare oneself against possible risks ability.
Description
Technical field
The ranking method for the individual customer overdue loan grade based on multi-model that the present invention relates to a kind of.
Background technique
Financing lease business flourished in recent years, and financial service audient's scale of construction is increased sharply, but continued in current financial supervision
Under the overall situation that tightening, fund lever constantly rise, there is many thorny problems: collage-credit data for Risk-warning and collection after loan
Degree of integration is low, informationization technology degree is insufficient, human cost is higher, risk identification degree is low, collection low efficiency, overdue loss are tight
Assets processing cost height etc. after weight, loan.
Summary of the invention
In view of the above-mentioned problems in the prior art, the main purpose of the present invention is to provide a kind of based on multi-model
The ranking method of individual customer overdue loan grade, for the processing of overdue client's differentiation, preferentially to the grading higher visitor of risk
Family is handled, and is saved manpower, is promoted collection efficiency and customer experience, prepare oneself against possible risks ability.
The technical scheme is that such:
A kind of ranking method of the individual customer overdue loan grade based on multi-model, comprising the following steps:
In client's overdue 7 days, 14 days, 21 days and 30 days periods, acquisition client is overdue each overdue stage
Feature, building violation correction model corresponding with each overdue stage respectively, and respectively M7 violation correction model, M14 break a contract
Prediction model, M21 violation correction model and M30 violation correction model；
M7 violation correction model, M14 violation correction model, M21 violation correction model and M30 based on above-mentioned building are disobeyed
About prediction model establishes the default risk grade for dividing each overdue stage, and the default risk grade is respectively A1 grades, A2
Grade, A3 grades, A4 grades and A5 grades；
The overdue feature of current overdue client is obtained, and according to the client in different overdue stages, calls corresponding promise breaking pre-
Survey model；
The risk of loss grade of current overdue client is constructed, and the risk of loss grade is respectively B1 grades, B2 grades, B3
Grade, B4 grades and B5 grades；
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained.
The overdue feature includes essential attribute feature, main strategies data, vehicle location information or client also amount of money
According to one of or it is a variety of.
Default risk grade corresponding to A1 grades described, A2 grades, A3 grades, A4 grades and A5 grades is respectively rudimentary promise breaking wind
Danger, intermediate default risk, more advanced default risk, advanced default risk and high grade default risk.
The overdue feature of current overdue client is obtained, and according to the client in different overdue stages, calls corresponding promise breaking pre-
Model is surveyed, specifically: the overdue client in overdue 7 days is without grading；Overdue 8~14 days overdue clients call M7 promise breaking
Prediction model；Overdue 15~21 days overdue clients call M14 violation correction model；Overdue 22~30 days overdue clients adjust
With M21 violation correction model；Overdue 31~60 days overdue clients call M30 violation correction model；Overdue 60 days or more exceed
Phase internal rating is high risk class.
Risk of loss grade corresponding to B1 grades described, B2 grades, B3 grades, B4 grades and B5 grades is respectively rudimentary risk damage
Mistake, intermediate risk of loss, more advanced risk of loss, advanced risk of loss and high grade risk of loss.
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 8~
14 days and 15~21 days overdue clients carry out according to following ranking method, in which: default risk grade is A5 or risk of loss
Grade is the overdue client of B5, and composite rating is X5 grades；Default risk grade is that A4 or risk of loss are rated the overdue of B4
Client, composite rating are X3 grades；Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B3 or B2,
Composite rating is X2 grades；Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B1, and composite rating is
X1 grades.
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 22
~30 days overdue clients carry out according to following ranking method, in which: default risk grade is that A5 or risk of loss are rated B5
Overdue client, composite rating be X5 grades；Default risk grade is that A4 or risk of loss are rated the overdue of B5 or B3 or B2
Client, composite rating are X5 grades；Default risk grade is the overdue client that A3 or A4 or risk of loss are rated B4, synthesis
It is rated X4 grades；Default risk grade is the overdue client that A2 and risk of loss are rated A3 or A2, and composite rating is X3 grades；
Default risk grade is the overdue client that A2 and risk of loss are rated B1, and composite rating is X2 grades.
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 31
~60 days overdue clients carry out according to following ranking method, in which: default risk grade is that A5 or risk of loss are rated B5
Overdue client, composite rating be X5 grades；Default risk grade is the overdue client that A4 or risk of loss are rated B3 or B2,
Its composite rating is X4 grades；Default risk grade is the overdue client that A3 or risk of loss are rated B2 or B1, composite rating
It is X3 grades.
Composite rating corresponding to X1 grades described, X2 grades, X3 grades, X4 grades and X5 grades is inferior grade, middle grade, higher etc.
Grade, high-grade and high grade.
The present invention have the following advantages and beneficial effects: the present invention provide it is a kind of based on multi-model individual customer loan exceed
The ranking method of phase grade, for overdue client in different overdue stages, the risk score grade being calculated based on model
With two dimension translocation sortings of risk of loss grade, client comprehensive grade, the processing of risk client collection differentiation are obtained；The present invention
The risk information that the different overdue time points of overdue client can be captured, promotes prediction accuracy, covers more high risk clients；Same hour hands
To overdue client's differentiation processing, preferentially to grading the higher client of risk handle, save manpower, promoted collection efficiency and
Customer experience, prepare oneself against possible risks ability；It grades in conjunction with the risk of loss amount of money, promotes the pass of the biggish client of the risk of loss amount of money
Note degree advantageously reduces monetary losses degree.
Detailed description of the invention
Fig. 1 is a kind of ranking method of the individual customer overdue loan grade based on multi-model provided in an embodiment of the present invention
In overdue 8-14 days and overdue 15-21 days overdue client composite rating schematic diagram.
Fig. 2 is a kind of ranking method of the individual customer overdue loan grade based on multi-model provided in an embodiment of the present invention
In overdue 22-30 days overdue client composite rating schematic diagram.
Fig. 3 is a kind of ranking method of the individual customer overdue loan grade based on multi-model provided in an embodiment of the present invention
In overdue 31-60 days overdue client composite rating schematic diagram.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.The present invention being usually described and illustrated herein in the accompanying drawings is implemented
The component of example can be arranged and be designed with a variety of different configurations.Therefore, below to the reality of the invention provided in the accompanying drawings
The detailed description for applying example is not intended to limit the range of claimed invention, but is merely representative of selected implementation of the invention
Example.Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts
Every other embodiment, shall fall within the protection scope of the present invention.
The present invention is further illustrated with specific embodiment below with reference to accompanying drawings.
It is as shown in Figure 1 to Figure 3: the grading side of the individual customer overdue loan grade based on multi-model of the embodiment of the present invention
Method, comprising the following steps: in client's overdue 7 days, 14 days, 21 days and 30 days periods, obtain client in each overdue rank
The overdue feature of section, building violation correction model corresponding with each overdue stage respectively, and respectively M7 violation correction model,
M14 violation correction model, M21 violation correction model and M30 violation correction model；M7 violation correction mould based on above-mentioned building
Type, M14 violation correction model, M21 violation correction model and M30 violation correction model are established and divide disobeying for each overdue stage
About risk class, and the default risk grade is respectively A1 grades, A2 grades, A3 grades, A4 grades and A5 grades；Obtain current overdue visitor
The overdue feature at family, and according to the client in different overdue stages, call corresponding violation correction model；Construct current overdue client
Risk of loss grade, and the risk of loss grade is respectively B1 grades, B2 grades, B3 grades, B4 grades and B5 grades；It is disobeyed based on described
About risk class and risk of loss grade, obtain the composite rating of overdue client.
The overdue feature includes essential attribute feature, main strategies data, vehicle location information or client also amount of money
According to one of or it is a variety of.
Default risk grade corresponding to A1 grades described, A2 grades, A3 grades, A4 grades and A5 grades is respectively rudimentary promise breaking wind
Danger, intermediate default risk, more advanced default risk, advanced default risk and high grade default risk.
The overdue feature of current overdue client is obtained, and according to the client in different overdue stages, calls corresponding promise breaking pre-
Model is surveyed, specifically: the overdue client in overdue 7 days is without grading；Overdue 8~14 days overdue clients call M7 promise breaking
Prediction model；Overdue 15~21 days overdue clients call M14 violation correction model；Overdue 22~30 days overdue clients adjust
With M21 violation correction model；Overdue 31~60 days overdue clients call M30 violation correction model；Overdue 60 days or more exceed
Phase internal rating is high risk class.
Risk of loss grade corresponding to B1 grades described, B2 grades, B3 grades, B4 grades and B5 grades is respectively rudimentary risk damage
Mistake, intermediate risk of loss, more advanced risk of loss, advanced risk of loss and high grade risk of loss.
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 8~
14 days and 15~21 days overdue clients carry out according to following ranking method, in which: default risk grade is A5 or risk of loss
Grade is the overdue client of B5, and composite rating is X5 grades；Default risk grade is that A4 or risk of loss are rated the overdue of B4
Client, composite rating are X3 grades；Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B3 or B2,
Composite rating is X2 grades；Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B1, and composite rating is
X1 grades.
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 22
~30 days overdue clients carry out according to following ranking method, in which: default risk grade is that A5 or risk of loss are rated B5
Overdue client, composite rating be X5 grades；Default risk grade is that A4 or risk of loss are rated the overdue of B5 or B3 or B2
Client, composite rating are X5 grades；Default risk grade is the overdue client that A3 or A4 or risk of loss are rated B4, synthesis
It is rated X4 grades；Default risk grade is the overdue client that A2 and risk of loss are rated A3 or A2, and composite rating is X3 grades；
Default risk grade is the overdue client that A2 and risk of loss are rated B1, and composite rating is X2 grades.
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 31
~60 days overdue clients carry out according to following ranking method, in which: default risk grade is that A5 or risk of loss are rated B5
Overdue client, composite rating be X5 grades；Default risk grade is the overdue client that A4 or risk of loss are rated B3 or B2,
Its composite rating is X4 grades；Default risk grade is the overdue client that A3 or risk of loss are rated B2 or B1, composite rating
It is X3 grades.
Composite rating corresponding to X1 grades above-mentioned, X2 grades, X3 grades, X4 grades and X5 grades is inferior grade, middle grade, higher etc.
Grade, high-grade and high grade.
The ranking method of the individual customer overdue loan grade based on multi-model of the embodiment of the present invention, can be applied to individual
Vehicle borrows the grading after loan, specifically includes the following steps:
(1) the customer default data of the overdue crucial time point of analysis client, most clients go back automatically within overdue 7 days
Money, whens overdue 14 days, 21 days, 30 days, overdue rate respectively increased by 10% amplitude, therefore obtained client overdue 7 days, 14 days, 21
It, 30 days when, essential attribute feature, main strategies data, vehicle location information, the client's refund data of client are counted
According to exceptional value, the processing of missing values, variable effectively and reasonably is screened, is trained using machine learning model, it will be trained
Four violation correction model Ms 7, M14, M21, M30 are deployed in system after loan, and wherein M7, M14, M21 use GBM integrated model, defeated
Result is the Default Probability of client overdue M3+ within 6 months futures out, and M30 is based on decision tree and generates several high risks rules；
(2) risk level standard for dividing each overdue stage is established based on aforementioned four violation correction model result,
The present invention mainly divides each risk class according to the practical promise breaking client accounting in each model score section and each section risk promotion degree,
And each risk class is respectively A1 grades, A2 grades, A3 grades, A4 grades and A5 grades, wherein A1 grades (low for rudimentary default risk
Grade), A2 grades are intermediate default risk (middle grade), and A3 grades are more advanced default risk (higher level), and A4 grades are advanced promise breaking
Risk (high-grade), A5 grades are high grade default risk (high grade), high grade promise breaking client's accounting equal 80% or more, high
Grade breaks a contract client's accounting 40%~55% or so, and higher level breaks a contract client's accounting 20%~30% or so, middle grade
The promise breaking client accounting of division is 10%~15%, and lower assessment graduation promise breaking client's accounting is below 8% or so；It considers simultaneously
The higher practical business form of the overdue longer customer risk of number of days, using staged grade classification mode, M7, M14 divide five layers
Risk class, M21 model result be divided into, higher, high, high grade, without the division of inferior grade, M30 model is only wrapped
Containing higher, high, high grading, guarantees that the more client of overdue number of days obtains higher attention rate in grading, promote grading
Interpretation, referring specifically to the following table 1.
The default risk distribution of grades in each overdue stage of table 1.
(3) currently the essential attribute feature, main strategies data, vehicle location information of overdue client, client is obtained to go back
Amount of money evidence calls corresponding model according to the client of different overdue time points, obtains model and return the result；Client in overdue 7 days
Without grading, overdue 8~14 days calling M7 violation correction models, overdue 15~21 days calling M14 violation correction models are overdue
22~30 days calling M21 violation correction models, overdue 31~60 days calling M30 violation correction models, overdue 60D+ internal rating
It is high risk class；
(4) the risk of loss amount of money for calculating current overdue client, i.e., if overdue customer default, the estimated damage occurred of enterprise
Lose the amount of money, the present invention in, the risk of loss amount of money is calculated by following formula: remaining sum-guarantee fund-vehicle residual value.Risk of loss
The amount of money be divided into it is low, in, higher, high, high five grades；
(5) default risk grade and risk of loss level crossings obtain overdue client comprehensive grading；Low default risk and low
Risk of loss is rated low-risk client, can wouldn't pay close attention to, most clients can self-healing, even if breaking a contract, in enterprise's band
The loss come is lower, and high default risk or high risk loss client need to take collection measure in time.
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 8~
14 days and 15~21 days overdue clients carry out according to following ranking method, in which:
The overdue client that default risk grade is A5 or risk of loss grade is B5, composite rating are high grade (X5
Grade)；
Default risk grade is the overdue client that A4 or risk of loss are rated B4, and composite rating is more advanced (X3
Grade)；
Default risk grade is high, and composite rating is advanced；
Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B3 or B2, and composite rating is medium
Grade (X2 grades)；
Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B1, and composite rating is inferior grade
(X1 grades).
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 22
~30 days overdue clients carry out according to following ranking method, in which:
Default risk grade is the overdue client that A5 or risk of loss are rated B5, and composite rating is high grade (X5
Grade)；
Default risk grade is the overdue client that A4 or risk of loss are rated B5 or B3 or B2, and composite rating is high
Grade (X5 grades)；
Default risk grade is the overdue client that A3 or A4 or risk of loss are rated B4, and composite rating is advanced (X4
Grade)；
Default risk grade is the overdue client that A2 and risk of loss are rated A3 or A2, and composite rating is more advanced
(X3 grades)；
Default risk grade is the overdue client that A2 and risk of loss are rated B1, and composite rating is middle grade (X2
Grade).
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 31
~60 days overdue clients carry out according to following ranking method, in which:
Default risk grade is the overdue client that A5 or risk of loss are rated B5, and composite rating is high grade (X5
Grade)；
Default risk grade is the overdue client that A4 or risk of loss are rated B3 or B2, and composite rating is advanced (X4
Grade)；
Default risk grade is the overdue client that A3 or risk of loss are rated B2 or B1, and composite rating is more advanced
(X3 grades).
A kind of ranking method of individual customer overdue loan grade based on multi-model provided in an embodiment of the present invention, is commented
The calculating logic of valence method is as follows:
(1) essential attribute feature, main strategies data, vehicle location information, the client's also amount of money of overdue client are obtained
According to etc., data preprocessing operation is carried out, violation correction model is constructed；
(2) to the overdue client of production system carry out promise breaking scoring calculating, use trained violation correction model M 7,
M14, M21, M30 calculate the different overdue overdue clients of time point, and the risk score of promise breaking (overdue M3+) within 6 months futures divides
Default risk grade；
(3) the risk of loss amount of money and risk of loss grade of client, risk score grade and wind based on each stage are calculated
Two dimension translocation sortings of dangerous loss, evaluate client comprehensive risk class, refine risk control, identify that self-healing client and height disobey
About risk, high loss client, cut operating costs.
A kind of ranking method of individual customer overdue loan grade based on multi-model provided in an embodiment of the present invention, tool
There are following characteristics:
1, overdue 7 days, 14 days, 21 days, excavate within 30 days the overdue feature of each overdue time, the integrated scoring mould of exploitation respectively
Type, current overdue client calls corresponding violation correction model to predict respectively, compared to the overdue client of full dose under traditional mode
The rough prediction mode of single model can more capture the risk information of the different overdue time points of overdue client, promote prediction accuracy,
Cover more high risk clients；
2, overdue client's differentiation processing, is preferentially handled the grading higher client of risk, is saved manpower, is promoted and urged
Rate of producing effects and customer experience, prepare oneself against possible risks ability；
3, it grades in conjunction with the risk of loss amount of money, promotes the attention rate of the biggish client of the risk of loss amount of money, advantageously reduce
Monetary losses degree.
Finally, it should be noted that above-described embodiments are merely to illustrate the technical scheme, rather than to it
Limitation；Although the present invention is described in detail referring to the foregoing embodiments, those skilled in the art should understand that:
It can still modify to technical solution documented by previous embodiment, or to part of or all technical features into
Row equivalent replacement；And these modifications or substitutions, it does not separate the essence of the corresponding technical solution various embodiments of the present invention technical side
The range of case.
Claims (9)
1. a kind of ranking method of the individual customer overdue loan grade based on multi-model, it is characterised in that: the following steps are included:
In client's overdue 7 days, 14 days, 21 days and 30 days periods, overdue feature of the client in each overdue stage is obtained,
Building violation correction model corresponding with each overdue stage respectively, and respectively M7 violation correction model, M14 violation correction mould
Type, M21 violation correction model and M30 violation correction model；
M7 violation correction model, M14 violation correction model, M21 violation correction model and M30 promise breaking based on above-mentioned building are pre-
Model is surveyed, establishes the default risk grade for dividing each overdue stage, and the default risk grade is respectively A1 grades, A2 grades, A3
Grade, A4 grades and A5 grades；
The overdue feature of current overdue client is obtained, and according to the client in different overdue stages, calls corresponding violation correction mould
Type；
The risk of loss grade of current overdue client is constructed, and the risk of loss grade is respectively B1 grades, B2 grades, B3 grades, B4
Grade and B5 grades；
Based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained.
2. the ranking method of the individual customer overdue loan grade according to claim 1 based on multi-model, feature exist
In the overdue feature includes in essential attribute feature, main strategies data, vehicle location information or client's refund data
It is one or more.
3. the ranking method of the individual customer overdue loan grade according to claim 1 based on multi-model, feature exist
In default risk grade corresponding to A1 grades described, A2 grades, A3 grades, A4 grades and A5 grades is respectively rudimentary default risk, middle rank
Default risk, more advanced default risk, advanced default risk and high grade default risk.
4. the ranking method of the individual customer overdue loan grade according to claim 1 based on multi-model, feature exist
In, the overdue feature of the current overdue client of acquisition, and according to the client in different overdue stages, call corresponding violation correction mould
Type, specifically:
Overdue client in overdue 7 days is without grading；
Overdue 8~14 days overdue clients call M7 violation correction model；
Overdue 15~21 days overdue clients call M14 violation correction model；
Overdue 22~30 days overdue clients call M21 violation correction model；
Overdue 31~60 days overdue clients call M30 violation correction model；
Overdue 60 days or more overdue internal ratings are high risk class.
5. the ranking method of the individual customer overdue loan grade according to claim 1 based on multi-model, feature exist
In risk of loss grade corresponding to B1 grades described, B2 grades, B3 grades, B4 grades and B5 grades is respectively rudimentary risk of loss, middle rank
Risk of loss, more advanced risk of loss, advanced risk of loss and high grade risk of loss.
6. the ranking method of the individual customer overdue loan grade according to claim 1 based on multi-model, feature exist
In, be based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 8~14 days
With 15~21 days overdue clients, carried out according to following ranking method, in which:
The overdue client that default risk grade is A5 or risk of loss grade is B5, composite rating are X5 grades；
Default risk grade is the overdue client that A4 or risk of loss are rated B4, and composite rating is X3 grades；
Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B3 or B2, and composite rating is X2 grades；
Default risk grade is the overdue client that A1 or A2 and risk of loss are rated B1, and composite rating is X1 grades.
7. the ranking method of the individual customer overdue loan grade according to claim 6 based on multi-model, feature exist
In, be based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 22~30 days
Overdue client, according to following ranking method carry out, in which:
Default risk grade is the overdue client that A5 or risk of loss are rated B5, and composite rating is X5 grades；
Default risk grade is the overdue client that A4 or risk of loss are rated B5 or B3 or B2, and composite rating is X5 grades；
Default risk grade is the overdue client that A3 or A4 or risk of loss are rated B4, and composite rating is X4 grades；
Default risk grade is the overdue client that A2 and risk of loss are rated A3 or A2, and composite rating is X3 grades；
Default risk grade is the overdue client that A2 and risk of loss are rated B1, and composite rating is X2 grades.
8. the ranking method of the individual customer overdue loan grade according to claim 7 based on multi-model, feature exist
In, be based on the default risk grade and risk of loss grade, the composite rating of overdue client is obtained, for overdue 31~60 days
Overdue client, according to following ranking method carry out, in which:
Default risk grade is the overdue client that A5 or risk of loss are rated B5, and composite rating is X5 grades；
Default risk grade is the overdue client that A4 or risk of loss are rated B3 or B2, and composite rating is X4 grades；
Default risk grade is the overdue client that A3 or risk of loss are rated B2 or B1, and composite rating is X3 grades.
9. the ranking method of the individual customer overdue loan grade according to claim 8 based on multi-model, feature exist
In composite rating corresponding to X1 grades described, X2 grades, X3 grades, X4 grades and X5 grades is inferior grade, middle grade, higher level, height
Grade and high grade.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910490288.8A CN110533520A (en) | 2019-06-06 | 2019-06-06 | A kind of ranking method of the individual customer overdue loan grade based on multi-model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910490288.8A CN110533520A (en) | 2019-06-06 | 2019-06-06 | A kind of ranking method of the individual customer overdue loan grade based on multi-model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110533520A true CN110533520A (en) | 2019-12-03 |
Family
ID=68659352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910490288.8A Pending CN110533520A (en) | 2019-06-06 | 2019-06-06 | A kind of ranking method of the individual customer overdue loan grade based on multi-model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110533520A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199477A (en) * | 2019-12-20 | 2020-05-26 | 北京淇瑀信息科技有限公司 | Multi-stage hybrid risk management method and device and electronic equipment |
| CN111353901A (en) * | 2020-02-24 | 2020-06-30 | 阳光保险集团股份有限公司 | Risk identification monitoring method and device and electronic equipment |
| CN111784506A (en) * | 2020-07-01 | 2020-10-16 | 深圳前海微众银行股份有限公司 | Overdue risk control method, device and readable storage medium |
| CN112001790A (en) * | 2020-08-24 | 2020-11-27 | 中国银行股份有限公司 | Block chain-based post-loan monitoring processing method and system and nodes |
| CN112017059A (en) * | 2020-07-14 | 2020-12-01 | 北京淇瑀信息科技有限公司 | Hierarchical optimization risk control method and device and electronic equipment |
| CN112767121A (en) * | 2020-12-31 | 2021-05-07 | 山东数字能源交易中心有限公司 | Method and device for processing risk level data |
| CN113450207A (en) * | 2021-06-29 | 2021-09-28 | 重庆农村商业银行股份有限公司 | Intelligent collection accelerating method, device, equipment and storage medium |
| CN113538125A (en) * | 2021-06-29 | 2021-10-22 | 百维金科(上海)信息科技有限公司 | Risk rating method for optimizing Hopfield neural network based on firefly algorithm |
| CN113554227A (en) * | 2021-07-23 | 2021-10-26 | 中信银行股份有限公司 | Training method of user loss rate prediction model and user loss rate prediction method |
| CN113628026A (en) * | 2021-06-30 | 2021-11-09 | 重庆度小满优扬科技有限公司 | Method and device for predicting overdue risk ranking |
| CN114841802A (en) * | 2022-07-04 | 2022-08-02 | 天津金城银行股份有限公司 | Method and device for generating combined cycle credit wind control strategy |
| CN116071103A (en) * | 2023-03-07 | 2023-05-05 | 天津金城银行股份有限公司 | Method and device for prompting client to borrow and electronic equipment |
| CN117372153A (en) * | 2023-10-31 | 2024-01-09 | 金扁担(北京)数字科技有限公司 | Fraud risk model and credit risk model based collection promoting method |
-
2019
- 2019-06-06 CN CN201910490288.8A patent/CN110533520A/en active Pending
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199477A (en) * | 2019-12-20 | 2020-05-26 | 北京淇瑀信息科技有限公司 | Multi-stage hybrid risk management method and device and electronic equipment |
| CN111353901A (en) * | 2020-02-24 | 2020-06-30 | 阳光保险集团股份有限公司 | Risk identification monitoring method and device and electronic equipment |
| CN111784506A (en) * | 2020-07-01 | 2020-10-16 | 深圳前海微众银行股份有限公司 | Overdue risk control method, device and readable storage medium |
| CN112017059A (en) * | 2020-07-14 | 2020-12-01 | 北京淇瑀信息科技有限公司 | Hierarchical optimization risk control method and device and electronic equipment |
| CN112001790A (en) * | 2020-08-24 | 2020-11-27 | 中国银行股份有限公司 | Block chain-based post-loan monitoring processing method and system and nodes |
| CN112767121A (en) * | 2020-12-31 | 2021-05-07 | 山东数字能源交易中心有限公司 | Method and device for processing risk level data |
| CN113450207A (en) * | 2021-06-29 | 2021-09-28 | 重庆农村商业银行股份有限公司 | Intelligent collection accelerating method, device, equipment and storage medium |
| CN113538125A (en) * | 2021-06-29 | 2021-10-22 | 百维金科(上海)信息科技有限公司 | Risk rating method for optimizing Hopfield neural network based on firefly algorithm |
| CN113628026A (en) * | 2021-06-30 | 2021-11-09 | 重庆度小满优扬科技有限公司 | Method and device for predicting overdue risk ranking |
| CN113554227A (en) * | 2021-07-23 | 2021-10-26 | 中信银行股份有限公司 | Training method of user loss rate prediction model and user loss rate prediction method |
| CN114841802A (en) * | 2022-07-04 | 2022-08-02 | 天津金城银行股份有限公司 | Method and device for generating combined cycle credit wind control strategy |
| CN116071103A (en) * | 2023-03-07 | 2023-05-05 | 天津金城银行股份有限公司 | Method and device for prompting client to borrow and electronic equipment |
| CN117372153A (en) * | 2023-10-31 | 2024-01-09 | 金扁担(北京)数字科技有限公司 | Fraud risk model and credit risk model based collection promoting method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110533520A (en) | A kind of ranking method of the individual customer overdue loan grade based on multi-model | |
| Schiller | Global supply-chain networks and corporate social responsibility | |
| Fan et al. | The law of one price: evidence from the transitional economy of China | |
| Nayyar et al. | International trade and the agricultural sector in India | |
| Cubbage et al. | Global timber investments, wood costs, regulation, and risk | |
| CN106780140A (en) | Electric power credit assessment method based on big data | |
| CN108335216A (en) | A kind of insurance risk assessment method, apparatus, terminal device and storage medium | |
| CN108665165B (en) | Enterprise tax analysis system based on big data | |
| CN105719045A (en) | Retention risk determiner | |
| CN109214915A (en) | Borrow risk methods of marking, system and computer readable storage medium | |
| CN111027836A (en) | Enterprise public credit rating system and method | |
| CN104036367A (en) | Intelligent customer credit optimizing processing method based on ERP system | |
| KR20190130535A (en) | Pair Trading System based on the Stock bigdata and Controlling Method for the Same | |
| Athukorala | Manufacturing exports from Sri Lanka: opportunities, achievements and policy options | |
| CN106651582A (en) | Security transaction risk early warning method and system based on instruction stream toxicity index | |
| CN113869700A (en) | Performance index prediction method and device, electronic equipment and storage medium | |
| CN106981022A (en) | The economic indicator method for early warning and system of a kind of real estate project clearance | |
| CN110245879A (en) | A kind of risk rating method and device | |
| Li et al. | China’s export evolution in the dynamic global product space from 2000 to 2011 | |
| Portugal-Perez et al. | Why trade facilitation matters to Africa? | |
| Bems et al. | Economic Consequences of Large Extraction Declines: Lessons for the Green Transition | |
| CN114757596A (en) | Power business monitoring and early warning method and system and readable storage medium | |
| Winseck | Growth and Upheaval in the Network Media Economy, 1984-2021 | |
| Kyvik Nordås | The WTO Reference Paper meets EU common regulatory policy in CETA | |
| CN107665433A (en) | Method, apparatus, computer equipment and the storage medium of single reimbursement are removed in control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191203 |
|
| WD01 | Invention patent application deemed withdrawn after publication |