Systems and methods for income scoring

• US 7,451,095 B1
• Filed: 10/30/2002
• Issued: 11/11/2008
• Est. Priority Date: 10/30/2002
• Status: Active Grant

• Current Assignee: Freddie Mac
• Original Assignee: Freddie Mac
• Inventors: Gordon, J. Douglas, Sims, Pamela W., Bradley, Michael G.
• Primary Examiner(s): Vig; Naresh
• Application Number:
  US10/283,339
• Time in Patent Office: 2,204 Days
• Field of Search: 705/36, 705/1, 705/38
• US Class Current: 705/35
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• 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

Systems and methods consistent with the present invention provide a score indicating that a stated income is likely to be faulty. In one embodiment, a method includes receiving information representative of the stated income and determining the score based on the received information and a model, such that the score indicates a likelihood that the stated income is faulty.

Claims

1. A computer implemented method for providing a score indicating whether that a stated income of a borrower is likely to be faulty, said method comprising the steps of:

receiving by a computer

borrower information representative of the borrower from a first source, the borrower information including the stated income of the borrower;

based on the borrower information, income-verifying information from a second source;

said computer providing the score by;

initializing one or more variables of a model based on the stated income and the income-verifying information;

calculating one or more coefficients for the one or more initialized variables;

determining the score by multiplying the one or more initialized variables by the corresponding coefficient and summing the results, such that the score indicates whether the stated income of the borrower is likely to be faulty; and

providing the determined score.

2. The method of claim 1, wherein said step of determining further comprises the step of:

scaling the score into a range of score values, such that a low score value within the range of score values corresponds to a high likelihood that the stated income is faulty and a high score value within the range of score values corresponds to a low likelihood that the stated income is faulty.

3. The method of claim 2, wherein said step of scaling further comprises the step of:

scaling the score into a predetermined range of about 300 to about 900.

4. The method of claim 3, further comprising the step of:

scaling the score such that a difference of sixty represents twice the likelihood that the stated income is faulty by calculating actual coefficients of the model using the equation, actual coefficient=initial coefficient*(60/ln(2)), wherein the initial coefficient is determined using received historical information or a statistical technique.

5. The method of claim 1, wherein said step of determining further comprises the step of:

scaling the score into a range of score values, such that a low score value within the range of score values corresponds to a low likelihood that the stated income is faulty and a high score value within the range of score values corresponds to a high likelihood that the stated income is faulty.

6. The method of claim 1, wherein said step of receiving borrower information further comprises the step of:

receiving an identity of the borrower.

7. Method of claim 1, further comprising the step of:

defining the model as a mathematical model that determines the score as a function of one or more inputs.

8. The method of claim 1

wherein the coefficients are calculated based on a statistical technique for incorporation into the model.

9. The method of claim 8, further comprising the step of:

defining the statistical technique as a logistic regression.

10. The method of claim 1, further comprising the step of:

determining a combined score to indicate the likelihood that a combined income of a plurality borrowers is faulty.

11. The method of claim 10, further comprising the step of:

determining the combined score as a sum of the scores of the plurality of borrowers.

12. The method of claim 1, wherein

the stated income is received from the borrower.

13. The method of claim 1, wherein

the stated income is received from an entity other than the borrower.

14. A computerized system for providing a score indicating whether that a stated income of a borrower is likely to be faulty, said system comprising:

a processor;

means for receiving borrower information representative of the borrower from a first source, the borrower information including the stated income of the borrower;

means for receiving, based on the borrower information, income-verifying information from a second source;

means for initializing one or more variables of a model based on the stated income and the income-verifying information;

means for calculating one or more coefficients for the one or more initialized variables;

means for determining the score by multiplying the one or more initialized variables by the corresponding coefficient and summing the results, such that the score indicates whether the stated income of the borrower is likely to be faulty; and

means for providing the determined score.

15. The system of claim 14, wherein said means for determining further comprises:

means for scaling the score into a range of score values, such that a low score value within the range of score values corresponds to a high likelihood that the stated income is faulty and a high score value within the range of score values corresponds to a low likelihood that the stated income is faulty.

16. A computer program product for providing a score indicating whether a stated income of a borrower is likely to be faulty, the computer program product being stored in a computer-readable medium and comprising:

code for receiving borrower information representative of the borrower from a first source, the borrower information including the stated income of the borrower;

code for receiving, based on the borrower information, income-verifying information from a second source;

code for initializing one or more variables of a model based on the stated income and the income-verifying information;

code for calculating one or more coefficients for the one or more initialized variables;

code for determining the score by multiplying the one or more initialized variables by the corresponding coefficient and summing the results, such that the score indicates whether the stated income of the borrower is likely to be is faulty; and

code for providing the determined score;

wherein said plurality of codes are computer executable code.

17. The system of claim 16, wherein said code further comprises:

code that scales the score into a range of score values, such that a low score value within the range of score values corresponds to a high likelihood that the stated income is faulty and a high score value within the range of score values corresponds to a low likelihood that the stated income is faulty.

18. A network including a plurality of financial entities interconnected by a communication channel, wherein a first one of the financial entities provides a score indicating whether a stated income of a borrower is likely to be faulty, the first said financial entity comprising:

a computer program product stored in a computer-readable medium, the computer program product comprising;

code for receiving borrower information representative of the borrower from a first source, the borrower information including the stated income of the borrower;

code for receiving, based on the borrower information, income-verifying information from a second source;

code for initializing one or more variables of a model based on the stated income and the income-verifying information;

code for calculating one or more coefficients for the one or more initialized variables;

code for determining the score by multiplying the one or more initialized variables by the corresponding coefficient and summing the results, such that the score indicates whether the stated income is likely to be faulty;

at least one processor for executing the code

wherein said plurality of codes are computer executable code.

19. A computer implemented method for determining a score indicating whether a stated income of a borrower is likely to be faulty, the method comprising:

receiving by a computer

borrower information representative of the borrower from a first source, the borrower information including the stated income of the borrower;

based on the borrower information, income-verifying information from a second source;

said computer providing the score by;

initializing one or more variables of a model based on the stated income and the income-verifying information;

calculating one or more coefficients for the one or more initialized variables;

determining and providing the score by multiplying the one or more initialized variables by the corresponding coefficient and summing the results, such that the score indicates whether the stated income of the borrower is likely to be faulty.

20. A system comprising a processor configured to execute

code that receives borrower information representative of the borrower from a first source, the borrower information including the stated income of the borrower;

code that receives, based on the borrower information, income-verifying information from a second source;

code that initializes one or more variables of a model based on the stated income and the income-verifying information;

code that calculates one or more coefficients for the one or more initialized variables;

code that determines and provides a score by multiplying the one or more initialized variables by the corresponding coefficient and summing the results, such that the score indicates whether the stated income of the borrower is likely to be faulty, and

wherein said plurality of codes are computer executable code.

21. The system of claim 20, wherein the at least one processor interfaces with a communication channel.

22. The system of claim 21, wherein the communication channel is the Internet.

23. The system of claim 20, wherein said memory further comprises:

code that determines the score based on a statistical technique.

24. The system of claim 23, wherein the statistical technique is a logistic regression.