Methods and systems of predicting mortgage payment risk

• US 8,065,234 B2
• Filed: 06/16/2011
• Issued: 11/22/2011
• Est. Priority Date: 03/24/2006
• Status: Active Grant

• Current Assignee: Corelogic Solutions LLC (CoreLogic Incorporated)
• Original Assignee: CoreLogic Information Solutions Incorporated (CoreLogic Incorporated)
• Inventors: Liao, Yuansong, Yan, Rui
• Primary Examiner(s): Colbert; Ella
• Assistant Examiner(s): Tran; Hai
• Application Number:
  US13/162,517
• Publication Number:
  US 20110251945A1
• Time in Patent Office: 159 Days
• Field of Search: 705 35- 45
• US Class Current: 705/44
• CPC Class Codes: G06Q 20/10 specially adapted for electronic funds transfer [EFT] systems; specially adapted for home banking systems
• CPC Class Codes: G06Q 20/40 Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof

Abstract

Disclosed herein are methods and systems for detecting a risk of payment default. In various embodiments, the systems/methods receive mortgage data associated with a mortgage application of an applicant, generate one or more models based on data related to historical mortgage transactions and determine a payment default risk score based at least partly on the one or more generated models, the mortgage data associated with the mortgage application, and the credit data related to the applicant. In another embodiment, a system is disclosed for selecting mortgage applications for further fraud evaluation based on the determined risks of early payment default associated with the mortgage applications and performing the fraud evaluation on those selected mortgage applications.

Claims

1. A system for detecting a risk of payment defaults, comprising:

non-transitory data storage that stores (1) mortgage data associated with a mortgage application that is associated with an applicant and (2) credit data related to the applicant; and

a computer system comprising one or more processors, the computer system being coupled with the data storage and being configured to;

process model training data related to a plurality of historical mortgage transactions;

generate one or more models based at least partly on the model training data by a process that comprises at least one of;

identifying, within the plurality of historical mortgage transactions, historical mortgage transactions that resulted in a payment default and historical mortgage transactions that did not result in a payment default; and

identifying input variables that are predictive of payment default based on the data related to the historical mortgage transactions; and

determine a payment default risk score for the mortgage application based at least partly on (1) the one or more generated models, (2) the mortgage data associated with the mortgage application, and (3) the credit data related to the applicant.

2. The system of claim 1, wherein the computer system is further configured to generate report data indicative of a risk of payment default based at least partly on the payment default risk score.

3. The system of claim 2, wherein the computer system is further configured to generate the report data indicative of a risk of payment default by at least matching a plurality of high risk rules to the received mortgage data.

4. The system of claim 3, wherein the high risk rules are generated based on data related to historical mortgage transactions that resulted in a payment default.

5. The system of claim 4, wherein the report data comprises at least one risk indicator of payment default generated based on result of the high risk rules matching.

6. The system of claim 5, wherein the at least one risk indicator include a risk indicator reflective of at least one of a borrower'"'"'s risk, a borrower'"'"'s affordability risk, a property valuation risk, a geographic risk, or a combination thereof.

7. The system of claim 2, wherein the generated report data reflects a risk of early payment default occurring within a time period that is shorter than or equal to 18 months of a funding of a loan sought for by the mortgage application.

8. The system of claim 1, wherein the payment default risk score is reflective of one or more of:

a risk related to the applicant'"'"'s credit data;

a risk related to an appraised value of a subject property in the mortgage application; and

a risk related to the geographic area in which the subject property is located.

9. The system of claim 1, wherein the computer system is further configured to generate the one or more models based on a supervised model training technique.

10. The system of claim 1, wherein the model training data is segmented geographically to account for regional differences in home prices and lending practices.

11. The system of claim 1, wherein the computer system is configured to generate the one or more models by a process that comprises identifying, within the plurality of historical mortgage transactions, historical mortgage transactions that resulted in a payment default and historical mortgage transactions that did not result in a payment default.

12. The system of claim 1, wherein the computer system is configured to generate the one or more models by a process that comprises identifying input variables that are predictive of payment default based on the data related to the historical mortgage transactions.

13. A method for detecting a risk of payment default, comprising:

generating one or more models based at least partly on model training data related to a plurality of historical mortgage transactions, the generating comprising at least one of;

identifying, within the plurality of historical mortgage transactions, historical mortgage transactions that resulted in a payment default and historical mortgage transactions that did not result in a payment default; and

identifying input variables that are predictive of payment default based on the data related to the historical mortgage transactions;

receiving mortgage data associated with a mortgage application that is associated with an applicant and credit data related to the applicant; and

determining a payment default risk score for the mortgage application based at least partly on the one or more generated models, the mortgage data associated with the mortgage application, and the credit data related to the applicant,

wherein the method is performed in its entirety by a computing system that comprises one or more computing devices.

14. The method of claim 13, further comprising generating report data indicative of a risk of payment default based at least partly on the payment default risk score.

15. The method of claim 14, wherein the high risk rules are generated based on data related to mortgage transactions that resulted in a payment default.

16. The method of claim 15, the report data comprises at least one risk indicator of payment default generated based on result of the high risk rules matching.

17. The method of claim 16, wherein the at least one risk indicator include a risk indicator reflective of at least one of a borrower'"'"'s risk, a borrower'"'"'s affordability risk, a property valuation risk, a geographic risk, or a combination thereof.

18. The method of claim 14, wherein the generated report data reflects a risk of early payment default occurring within a time period that is shorter than or equal to 18 months of a funding of a loan sought for by the mortgage application.

19. The method of claim 13, wherein the payment default risk score is reflective of one or more of:

a risk related to the applicant'"'"'s credit data;

a risk related to an appraised value of a subject property in the mortgage application; and

a risk related to the geographic area in which the subject property is located.

20. The method of claim 13, wherein generating the one or more models is based on a supervised model training technique.

21. The method of claim 13, wherein the model training data is segmented geographically to account for regional differences in home prices and lending practices.

22. The method of claim 13, wherein generating the report data indicative of a risk of payment default further comprises:

matching a plurality of high risk rules to the received mortgage data.

23. The method of claim 13, wherein generating the one or more models comprises identifying, within the plurality of historical mortgage transactions, historical mortgage transactions that resulted in a payment default and historical mortgage transactions that did not result in a payment default.

24. The method of claim 13, wherein generating the one or more models comprises identifying input variables that are predictive of payment default based on the data related to the historical mortgage transactions.