SYSTEMS AND METHODS FOR ANALYZING DATA

• US 20070214076A1
• Filed: 09/27/2006
• Published: 09/13/2007
• Est. Priority Date: 03/10/2006
• Status: Active Grant

• Current Assignee: Experian Information Solutions Incorporated (Experian plc)
• Original Assignee: Experian-Scorex, LLC (Experian plc)
• Inventors: Wang, Chien-Wei, Robida, Chuck
• Granted Patent: US 7,711,636 B2
• Time in Patent Office: Days
• US Class Current: 705/38
• CPC Class Codes: G06Q 20/10 specially adapted for electronic funds transfer [EFT] systems; specially adapted for home banking systems
• CPC Class Codes: G06Q 30/0204 Market segmentation
• 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
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

Information regarding individuals that fit a bad performance definition, such as individuals that have previously defaulted on a financial instrument or have declared bankruptcy, is used to develop a model that is usable to determine whether an individual that does not fit the bad performance definition is more likely to subsequently default on a financial instrument or to declare bankruptcy. The model may be used to generate a score for each individual, and the score may be used to segment the individual into a segment of a segmentation structure that includes individuals with related scores, where segments may include different models for generating a final risk score for the individuals assigned to the particular segments. Thus, the segment to which an individual is assigned, which may be determined based at least partly on the score assigned to the individual, may affect the final risk score that is assigned to the individual.

Claims

1. A method of generating a default/bankruptcy model for assigning an individual to particular segments of a segmentation structure, wherein the default/bankruptcy model is indicative of an individual'"'"'s propensity to either default on one or more financial instruments or file for bankruptcy, the method comprising:

receiving observation data comprising financial and demographic information regarding a plurality of individuals, the observation data indicating characteristics of the individuals at an observation time;

receiving outcome data comprising financial and demographic information regarding the plurality of individuals fitting a bad performance definition, the outcome data indicating characteristics of the individuals fitting the bad performance definition during an outcome period, the outcome period beginning after the observation time; and

comparing the observation data and the outcome data in order to generate the bankruptcy/default model usable to determine which of a plurality of segments in the segmentation structure a particular individual should be assigned.

2. The method of claim 1, wherein the bad performance definition comprises individuals with at least one financial account that has previously had a ninety day past due status and individuals that have previously filed for bankruptcy.

3. The method of claim 1, wherein the outcome period is about 24 months.

4. The method of claim 1, wherein at least some of the segments are associated with respective final risk score models configured to generate final risk scores for each individual assigned to the respective segment.

5. The method of claim 1, wherein a bankruptcy/default score is determined for each individual using the bankruptcy/default model.

6. The method of claim 5, wherein the segmentation structure comprises at least two hierarchal levels of segments.

7. The method of claim 5, wherein the segmentation structure comprises at least three hierarchal levels of segments.

8. The method of claim 5, wherein the bankruptcy/default score is used to assign the individual to at least a final segment and a parent segment in the segmentation structure.

9. The method of claim 4, wherein a plurality of adverse action codes are associated with the particular individual, the adverse action codes each indicating a reason as to why the final risk score was assigned to the particular individual, wherein if being assigned to a particular segment affected the final risk score by a predetermined proportion, at least one of the plurality of adverse action codes relates to assignment to the particular segment.

10. A method of assessing a risk associated with an individual comprising:

generating a model based on data regarding a first subgroup of a population, the subgroup comprising a first portion fitting a first failure definition and a second portion fitting a second failure definition; and

applying the generated model to the individual, wherein the individual is not a member of the first subgroup.

11. The method of claim 10, wherein the first failure definition comprises individuals that have filed for bankruptcy and the second failure definition comprises individuals that have defaulted on a financial instrument.

12. The method of claim 10, wherein the generated model predicts whether the individuals is more likely to subsequently fit the first failure definition or to subsequently fit the second failure definition.

13. A computing system for segmenting each of a plurality of individuals into one of a plurality of segments of a segmentation structure, the system comprising:

a profile module configured to generate a default/bankruptcy model for assigning each individual to one or more segments of the segmentation structure, wherein the default/bankruptcy model is indicative of an individual'"'"'s propensity to either default on one or more financial instruments or to file for bankruptcy; and

a segmentation module configured to segment each of the individuals using the default/bankruptcy model, wherein the individuals include individuals satisfying a bad performance definition and individuals satisfying a good performance definition.

14. The system of claim 13, further comprising an adverse action module configured to associate a plurality of adverse action codes to each individual, the adverse action codes indicating reasons why a final risk score was assigned to a particular individual, wherein if being assigned to a particular segment affected a particular individual'"'"'s final risk score by a predetermined proportion at least one of the plurality of adverse action codes associated with the particular individual relates to assignment to the particular segment.

15. A method for selecting one or more adverse action codes to associate with a final risk score assigned to an individual, each of the adverse action codes indicating a reason that the final risk score was assigned to the individual, wherein the individual is assigned to a segmentation hierarchy comprising a plurality of segments, including a final segment, in a segmentation structure, the method comprising:

determining a first penalty associated with assignment of the individual to a final segment;

determining a first ratio of the first penalty to a difference between a highest possible final risk score and the final risk score for the individual;

if the determined first ratio is above a first determined threshold, allotting an adverse action code related to assignment of the individual to the final segment.

16. The method of claim 15, further comprising:

if the determined first ratio is above a second determined threshold, allotting another adverse action code related to assignment of the individual to the final segment.

17. The method of claim 15, further comprising:

determining a second penalty associated with assignment of the individual to a parent segment to the final segment;

determining a second ratio of the second penalty to a difference between a highest possible final risk score and the final risk score for the individual; and

if the determined second ratio is above the first determined threshold, allotting an adverse action code related to assignment of the individual to the parent segment.

18. The method of claim 15, wherein the final segment indicates the individual has a higher risk of filing for bankruptcy than defaulting on a financial account.

19. A method of generating a model for determining an individual'"'"'s propensity to enter either a first failure mode or a second failure mode, the method comprising:

defining a bad performance definition to include individuals that have characteristics of one or more of the first and second failure modes;

receiving observation data regarding a plurality of individuals fitting the bad performance definitions, the observation data indicating characteristics of the individuals at an observation time;

receiving outcome data regarding the plurality of individuals fitting the bad performance definition, the outcome data indicating characteristics of the individuals fitting the bad performance definition during an outcome period, the outcome period beginning after the observation time; and

comparing the observation data and the outcome data in order to generate a model usable to determine a likelihood that an individual not fitting the bad performance definition will enter a first failure mode or if the individual will enter the second failure mode.

20. The method of claim 19, wherein the first failure mode comprises filing for bankruptcy and the second failure mode comprises defaulting on a financial instrument.

21. The method of claim 19, wherein the first failure mode comprises defaulting on an installment loan and the second failure mode comprises defaulting on a revolving loan.

22. The method of claim 19, wherein the first failure mode comprises defaulting on a bank loan and the second failure mode comprises defaulting on an automobile loan.

23. The method of claim 19, wherein the observation time is about 24 months prior to generation of the model.

24. The method of claim 23, wherein the outcome period is a period of about 24 months prior to generation of the model, but exclusive of the observation time.