Systems, methods, and computer-readable storage media for analyzing HMDA data

• US 7,729,961 B1
• Filed: 04/14/2008
• Issued: 06/01/2010
• Est. Priority Date: 04/13/2007
• Status: Active Grant

• Current Assignee: Andrei Iancu, Ashley Moody Florida Office of The Attorney General and Successors Thereto, Jeanette Nunez Lieutenant Governor of Florida and Successors Thereto, Joan D. Heuer, Laurel M Lee Florida Secretary of State and Successors Thereto, Steven Mnuchin United States Secretary of The Treasury and Successors Thereto, Timothy E Gribben Commissioner For Bureau of The Fiscal Service Agency of The United States Department of The Treasury and Successors and Assigns Thereto
• Original Assignee: Federal Home Loan Mortgage Corporation
• Inventors: Gomez, Geni, Bickel, Keith
• Primary Examiner(s): Gart; Matthew S
• Assistant Examiner(s): PARIKH, HARSHAD R
• Application Number:
  US12/081,307
• Time in Patent Office: 778 Days
• Field of Search: 705/30
• US Class Current: 705/30
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03
• CPC Class Codes: G06Q 40/12 Accounting
• CPC Class Codes: G06Q 40/123 Tax preparation or submission

Abstract

Systems, methods, and computer-readable storage media are described for normalizing HMDA data. In one exemplary embodiment, a computer-implemented method of normalizing HMDA data comprises receiving HMDA data including at least one of HMDA data reports and loan-level public HMDA data, the HMDA data having information that varies as a function of time; correcting errors in the HMDA data; normalizing the HMDA data across any variation in the information; summarizing the normalized HMDA data; and outputting the summarized and normalized HMDA data to an application for analysis.

Claims

1. A computer-implemented method of normalizing public HMDA data received from a federal agency, the method comprising:

receiving, by a processor, Home Mortgage Disclosure Act (HMDA) data including at least one of HMDA data reports and loan-level public HMDA data, the HMDA data having loan information that varies as a function of time and that is provided by multiple financial institutions;

correcting, by the processor, errors in the HMDA data;

normalizing, by the processor, the HMDA data across variations in the information, wherein normalizing comprises;

accessing a first database storing data indicative of which HMDA data is provided by each of the multiple financial institutions,

accessing a second database storing the names of the multiple financial institutions and associated identifying tags that identify relationships among the multiple financial institutions,

reducing, based on the data stored in the first database and the names of the multiple financial institutions and the associated identifying tags stored in the second database, a total number of the names of the multiple financial institutions, and

reassigning, based on the reduced total number of the names of the multiple financial institutions, the data stored in the first database;

summarizing, by the processor, the normalized HMDA data; and

outputting, by the processor, the summarized and normalized HMDA data to an application for analysis.

2. The method of normalizing HMDA data according to claim 1, wherein correcting errors comprises correcting at least one of a missing census tract number and an improper data code.

3. The method of claim 1, wherein normalizing the HMDA data further comprises:

retrieving geographic reference data related to the HMDA data;

applying rules to the geographic reference data to normalize current census tract data to previous census tract data, current and previous census tract data to counties, and county information to metropolitan statistical area data; and

outputting normalized census tracts to metropolitan statistical area data.

4. The method of claim 3, wherein normalizing the HMDA data further comprises:

retrieving data related to a goal required of a government-sponsored enterprise and the normalized census tracts to metropolitan statistical area data;

applying rules to the retrieved data to determine loans within census tracts that meet the required goal; and

identifying loans that meet the required goal.

5. The method of claim 1, wherein normalizing the HMDA data further comprises:

applying rules to the accessed data indicative of which HMDA data is provided by each of the financial institutions to determine a market share of a first financial institution in a specified year and to include the market share of other financial institutions acquired by the first financial institution after the specified year.

6. The method of claim 1, wherein normalizing the HMDA data further comprises:

applying rules to the HMDA data to identify loans associated with the HMDA data belonging to the sub-prime market and loans belonging to the prime market.

7. The method of claim 1, wherein normalizing the HMDA data further comprises:

applying rules to the HMDA data to identify conforming loans or non-conforming loans.

8. A computer-readable storage medium including program instructions which, when executed by a processor, performs a method of normalizing public Home Mortgage Disclosure Act (HMDA) data received from a federal agency, the method comprising:

receiving HMDA data including at least one of HMDA data reports and loan-level public HMDA data, the HMDA data having loan information that varies as a function of time and that is provided by multiple financial institutions;

correcting errors in the HMDA data;

normalizing the HMDA data across variations in the information, wherein normalizing comprises;

accessing a first database storing data indicative of which HMDA data is provided by each of the multiple financial institutions,

accessing a second database storing names of the multiple financial institutions and associated identifying tags that identify relationships among the multiple financial institutions,

reducing, based on the data stored in the first database and the names of the multiple financial institutions and the associated identifying tags stored in the second database, a total number of the names of the multiple financial institutions, and

reassigning, based on the reduced total number of the names of the multiple financial institutions, the data stored in the first database;

summarizing the normalized HMDA data; and

outputting the summarized and normalized HMDA data to an application for analysis.

9. The storage medium of claim 8, wherein correcting errors comprises correcting at least one of a missing census tract number and an improper data code.

10. The storage medium of claim 8, wherein normalizing the HMDA data further comprises:

retrieving geographic reference data related to the HMDA data;

applying rules to the geographic reference data to normalize current census tract data to previous census tract data, current and previous census tract data to counties, and county information to metropolitan statistical area data; and

outputting normalized census tracts to metropolitan statistical area data.

11. The storage medium of claim 10, wherein normalizing the HMDA data further comprises:

retrieving data related to a goal required of a government-sponsored enterprise and the normalized census tracts to metropolitan statistical area data;

applying rules to the retrieved data to determine loans within census tracts that meet the required goal; and

identifying loans that meet the required goal.

12. The storage medium of claim 8, wherein normalizing the HMDA data further comprises:

applying rules to the accessed data indicative of which HMDA data is provided by each of the financial institutions to determine the market share of a first financial institution in a specified year and to include the market share of other financial institutions acquired by the first financial institution after the specified year.

13. The storage medium of claim 8, wherein normalizing the HMDA data further comprises:

applying rules to the HMDA data to identify loans associated with the HMDA data belonging to the sub-prime market and loans belonging to the prime market.

14. The storage medium of claim 8, wherein normalizing the HMDA data further comprises:

applying rules to the HMDA data to identify conforming loans or non-conforming loans.

15. A Home Mortgage Disclosure Act (HMDA) data analysis tool to perform a normalizing process and analysis of public HMDA data, received from a federal agency comprising:

a geographic translator to perform a cross-year process to normalize geographic information on locations of properties reported in the HMDA data provided by multiple financial institutions;

an income translator to perform a cross-year process to normalize HUD goals required of government-sponsored enterprises relative to the HMDA data;

a lender translator for to access a first database storing data indicative of which HDMA data is provided by each of the multiple financial institutions and a second database storing names of the multiple financial institutions and associated identifying tags that identify relationships among the multiple financial institutions, and to perform a cross-year process to normalize lender information reported in the HMDA data by;

reducing, based on the data stored in the first database and the names of the multiple financial institutions and the associated identifying tags stored in the second database, a total number of the names of the multiple financial institutions, and reassigning, based on the reduced total number of the names of the multiple financial institutions, the data stored in the first database;

a rules repository for storing rules for performing the normalization process; and

a rules processor for executing the rules.

16. The HMDA data analysis tool of claim 15, wherein the cross-year process performed by the geographic translator further comprises:

determining changes in HMDA reporting standards and formatting of the HMDA data, data definitions, and a time period when a change occurred; and

normalizing the HMDA data based on the determined change and the time period of the change.

17. The HMDA data analysis tool of claim 15, wherein the determined change is a change in census tract data.

18. The HMDA data analysis tool of claim 15, wherein the determined change is a change in a metropolitan statistical area definition.

19. The HMDA data analysis tool of claim 15, wherein the cross-year process performed by the income translator further comprises:

analyzing the HMDA data in relation to the HUD goals;

and flagging loans meeting the goals required of government-sponsored enterprises by the U.S. Department of Housing and Urban Development.

20. The HMDA data analysis tool of claim 15, wherein the cross-year process performed by the lender translator further comprises:

determining a market share of a financial institution for a particular year; and

analyzing the HMDA data in relation to the determined market share.