Automated method and system for predicting and/or verifying income
An income predictor deploys an income model that is based on credit information obtained for a representative group of the general population. The credit information is matched against actual employment and income information to predict an income range having a specified confidence factor. The income predictor can be used by subscribers (e.g., creditors, employers, etc.) to predict and/or verify the income of a candidate (e.g., applicants, potential employees, etc.).
1 . A method of predicting income, comprising:
creating an income model by evaluating a number of records that contain actual income information, wherein the number of records is a representative sample of an actual population;
accessing a credit report for a candidate; and
applying the income model to the credit report to generate a prediction of an income of the candidate, wherein the prediction is statistically significant.
2 . The method of claim 1 , further comprising:
sending the prediction to a third party for verification.
3 . The method of claim 2 , further comprising:
refining the income model if the prediction is not accurate based on the verification by the third party.
4 . A method of approving a loan, comprising:
predicting an income of a candidate, wherein the predicting comprises:
(a) creating an income model by evaluating a number of records that contain actual income information, wherein the number of records is representative of an actual population;
(b) accessing a credit report for the candidate; and
(c) applying the income model to the credit report to generate a prediction of the income of the candidate, wherein said prediction is statistically significant;
receiving an income statement from the candidate;
comparing the income statement with the prediction to determine whether a difference between the income statement and the prediction is within a predefined tolerance; and
verifying the income statement or the prediction via a third party if the difference is greater than the predefined tolerance.
5 . The method of claim 4 , further comprising:
taking no further action if the difference is within the predefined tolerance.
6 . A computer program product comprising a computer useable medium having computer readable program code functions embedded in said medium for causing a computer to predict income, comprising:
a first computer readable program code function that causes the computer to create an income model by evaluating a number of records that contain actual income information, wherein the number of records is representative of an actual population;
a second computer readable program code function that causes the computer to access a credit report for a candidate; and
a third computer readable program code function that causes the computer to apply the income model to the credit report to generate a prediction of an income of the candidate, wherein the prediction is statistically significant.
7 . The computer program product of claim 6 , further comprising:
a fourth computer readable program code function that causes the computer to send the prediction to a third party for verification.