METHOD AND APPARATUS SYSTEM FOR MODELING CONSUMER CAPACITY FOR FUTURE INCREMENTAL DEBT IN CREDIT SCORING

• US 20090037323A1
• Filed: 02/06/2008
• Published: 02/05/2009
• Est. Priority Date: 08/01/2007
• Status: Active Grant

• Current Assignee: Fair Isaac Corporation
• Original Assignee: Fair Isaac Corporation
• Inventors: Jack, Jennifer Elizabeth, FEINSTEIN, Jeffrey Allen, Sullivan, Gary J.
• Granted Patent: US 8,099,356 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 20/108 Remote banking, e.g. home banking
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis

Abstract

Predicting impact of future actions on subsequent creditworthiness involves developing a prediction model that predicts a statistical interaction of performance expectation with likely post-scoring behavior. Including sensitivity to new, post-scoring date credit behaviors in the analytic solution greatly improves snapshot score predictions. The modeling approach involves multiple snapshots: predictive and performance snapshots, plus an intermediate snapshot shortly after the predictive snapshot to quantify interim consumer behavior post-scoring date. Predictive interaction variables are calculated on the predictive data using simulated consumer profiles before and after assuming a sizeable simulated balance to infer the consumer'"'"'s tolerance for incremental future debt. Using an adjustor approach in predicting capacity allows isolation of the confounding effect of risk from the capacity determination. A resulting capacity index can be used to rank order originations and line increases according to capacity in consumer, bankcard, automobile and mortgage lending.

Claims

1. A computer-implemented method of predicting a consumer'"'"'s capacity for incremental debt, comprising the steps of:

storing computer-readable instructions which, when executed, implement a predictive model that simulates said consumer'"'"'s performance in view of additional debt incurred;

inputting data representing a credit risk score and weights for each of said plurality of prediction characteristics to said predictive model; and

outputting by said model a capacity index that estimates said consumer'"'"'s capacity to assume additional debt without default.

2. The method of claim 1, further comprising the step of reporting said capacity index accompanied by reason codes.

3. The method of claim 1, wherein categories of said prediction characteristics include:

an individual'"'"'s payment history;

amounts an individual owes on all accounts;

length of credit history;

new credit;

mix of credit types

total number of consumer'"'"'s revolving credit cards;

the consumer'"'"'s revolving balance; and

the consumer'"'"'s credit limit.

4. The method of claim 1, further comprising the step of generating said prediction characteristics from consumer-level credit data, said credit data comprising any of:

income;

custom scores;

master file variables;

credit bureau data; and

other third-party data.

5. The method of claim 1, wherein said predictive model comprises a scorecard model that ranks orders consumers according to said capacity.

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

generating said predictive model.

7. The method of claim 6, wherein said step of generating said predictive model comprises the steps of:

selecting a sample population of accounts from a source population;

for each account, generating a first snapshot comprising a credit risk score and said plurality of prediction characteristics for each member of said sample population at the beginning of a predetermined performance window;

for each account, determining good/bad performance behavior on all reported trade lines for duration of said performance window from a second snapshot generated at the end of said performance period;

for each account, weighting good/bad performance by incremental balance incurred during an intermediate period between said first and second snapshots as determined by a third snapshot;

for each account, generating a tag that expresses performance, given said incremental balance incurred;

for all accounts in aggregate, by means of a software modeling tool, creating a predictive model that describes the relationship among variables that rewards good performance and penalizes bad performance such that accounts showing a high capacity are ranked higher than accounts showing low capacity; and

for each account, adjusting said model by subtracting said risk score, wherein the remainder comprises said capacity index;

8. The method of claim 7, wherein said sample population is stratified, wherein one or more profile types likely to be under represented in a random sample are over sampled relative to said source population and wherein profile types are selectively weighted in a manner so that the weighted distribution of profile types mirrors the raw distribution of profile types in the source population.

9. A computer readable storage medium encoded with instructions, which when loaded into a digital computational device performs a method of predicting a consumer'"'"'s capacity for incremental debt comprising the steps of:

generating a credit risk score and a plurality of prediction characteristics;

inputting said credit risk score and weights for each of said plurality of prediction characteristics to a predictive model that simulates said consumer'"'"'s performance in view of additional debt incurred;

outputting a capacity index that estimates said consumer'"'"'s capacity to assume additional debt without default.

10. An apparatus for predicting a consumer'"'"'s capacity for incremental debt, comprising:

a computing device, said computing device comprising a processing component and a storage component; and

computer-readable instructions residing in said storage component which, when executed by said processing component instruct said processor to perform operations comprising;

generating a credit risk score and a plurality of prediction characteristics;

inputting said credit risk score and weights for each of said plurality of prediction characteristics to a predictive model that simulates said consumer'"'"'s performance in view of additional debt incurred; and

outputting a capacity index that estimates said consumer'"'"'s capacity to assume additional debt without default.

11. The apparatus of claim 10, said operations further comprising reporting said capacity index accompanied by reason codes.

12. The apparatus of claim 10, wherein categories of said prediction characteristics include:

an individual'"'"'s payment history;

amounts an individual owes on all accounts;

length of credit history;

new credit;

mix of credit types

total number of consumer'"'"'s revolving credit cards;

the consumer'"'"'s revolving balance; and

the consumer'"'"'s credit limit.

13. The apparatus of claim 10, said operations further comprising generating said prediction characteristics from consumer-level credit data, said credit data comprising any of:

income;

custom scores;

master file variables;

credit bureau data; and

other third-party data.

14. The apparatus of claim 10, wherein said predictive model comprises a scorecard model that ranks orders consumers according to said capacity.

15. The apparatus of claim 10, said operations further comprising:

generating said predictive model.

16. The apparatus of claim 15, wherein generating said predictive model comprises:

selecting a sample population of accounts from a source population;

for each account, generating a first snapshot comprising a credit risk score and said plurality of prediction characteristics for each member of said sample population at the beginning of a predetermined performance window;

for each account, weighting good/bad performance by incremental balance incurred during an intermediate period between said first and second snapshots as determined by a third snapshot;

for each account, generating a tag that expresses performance at the third snapshot, given said incremental balance incurred;

for all accounts, by means of a software modeling tool, creating a predictive model that describes the relationship among variables that rewards good performance and penalizes bad performance; and

for each account, adjusting said prediction model by subtracting said risk score, wherein the remainder comprises said capacity index such that accounts showing a high capacity are ranked higher than accounts showing low capacity.

17. The apparatus of claim 16, wherein said sample population is stratified, wherein one or more profile types likely to be under represented in a random sample are over sampled relative to said source population and wherein profile types are selectively weighted in a manner so that the weighted distribution of profile types mirrors the raw distribution of profile types in the source population.