Method of and apparatus for administering an asset-backed security using coupled lattice efficiency analysis

• US 7,574,396 B2
• Filed: 12/04/2002
• Issued: 08/11/2009
• Est. Priority Date: 12/04/2001
• Status: Active Grant

• Current Assignee: ICE Data Services, Inc. (Intercontinental Exchange, Inc.)
• Original Assignee: Andrew Kalotay Associates, Inc.
• Inventors: Kalotay, Andrew J., Yang, Deane
• Primary Examiner(s): Poinvil; Frantzy
• Assistant Examiner(s): GRAHAM, CLEMENT B
• Application Number:
  US10/309,109
• Publication Number:
  US 20030105696A1
• Time in Patent Office: 2,442 Days
• Field of Search: 705/36, 705/37, 705/34, 705/35, 705/38
• US Class Current: 705/37
• 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

Abstract

New tools for modeling prepayment of principal by obligors behind asset-backed securities are provided. The obligors are categorized into groups for ordered removal from the modeling and valuation process according to their refinancing efficiency, for instance. These tools can be used in a coupled-lattice recursive analysis process to obtain the value of asset-backed securities such as mortgage-backed securities and collateralized mortgage obligations.

Claims

1. A method for estimating the fair market value of an asset-backed security, said method being implemented in a computer having a processor and a computer-readable medium that stores program instructions, and comprising the steps of:

a. retrieving specifications of aspects of a plurality of assets, said assets generating cashflows from payment by obligors of principal and interest, wherein a level of the cashflows generated from said assets changes due to prepayments by said obligors;

b. retrieving a specification of the asset-backed security created from said plurality of assets;

c. determining, by execution of the program instructions in the processor, from said specifications of said aspects of the plurality of assets and said specification of said asset-backed security, a schedule of future cashflows that will be generated from the assets based on at least one of principal payments and interest payments from said obligors whose assets are members of said asset pool;

d. modeling, by execution of the program instructions in the processor, future prepayment behavior of said assets by said obligors, said modeling step including the sub-steps of;

i. defining at least three refinancing efficiency categories of obligors according to an expected propensity of the obligors to refinance the assets,

ii. partitioning the assets into asset pool efficiency buckets according to said refinancing efficiency categories,

iii. removing scheduled payments and core prepayments of principal which are not related to changing interest rates pro-rata from all asset pool efficiency buckets,

iv. attributing any excess of actual over scheduled prepayments to obligors refinancing assets in said asset pool, and

v. removing assets from said asset pool in a preset order according to the defined refinancing efficiency categories of obligors to reflect refinancing of assets by obligors and thus obtain a model of the remaining pool; and

e. determining, by execution of the program instructions in the processor, the current value of the asset-backed security, said current value determining step including the sub-steps of;

i. determining a core payment rate cash flow for said asset pool,

ii. generating a dual-coupled interest rate lattice from an interest rate environment information, including interest rates and respective interest rate volatilities, said lattice including a plurality of nodes representing points in time,

iii. determining, for the plurality of nodes in the lattice, whether each defined refinancing category of said asset pool would refinance based on the interest rate environment information,

iv. determining, for the plurality of nodes in the in the lattice, a respective first plurality of net cash flows for the asset-backed security, the first plurality of net cash flows including principal payments remaining in the asset pool after removal of assets prepaid due to refinancing,

v. determining, for the plurality of nodes in the in the lattice, a respective second plurality of net cash flows for the asset-backed security, the second plurality of cash flows including both principal and interest payments, and

vi. determining a value for said asset-backed security by discounting and adding the first and second plurality of net cash flows determined for the plurality of nodes in the lattice.

2. The computer-implemented method recited in claim 1, wherein said asset-backed securities are mortgage-backed securities and said obligors are mortgagors.

3. The computer-implemented method recited in claim 1, wherein said asset-backed securities are collateralized mortgage obligations and said obligors are mortgagors.

4. The computer-implemented method recited in claim 1, wherein said refinancing buckets of obligors include leapers who refinance before an optimal time, financial engineers who refinance at the optimal time, and laggards who refinance after the optimal time.

5. The computer-implemented method of claim 1, wherein said refinancing efficiency of obligors in deciding when to refinance is determined by a savings/option value, said saving/option value being calculated assuming that interest rates for similar assets will follow a random walk.

6. The computer-implemented method recited in claim 1, wherein the method sub-step of generating a dual coupled interest rate lattice from input interest rate environment includes the steps of:

i. modeling interest rates assuming correlated asset and asset-backed security yield curves which follow a random walk; and

ii. using said correlated asset and asset-backed security yield curves to construct said coupled asset-backed security/asset lattice.

7. The computer-implemented method recited in claim 1, further comprising:

(f) using said value for said asset-backed security in one or more activities selected from a group consisting of a portfolio of whole loans, managing a portfolio of mortgage service rights, and formulating portfolio strategies.

8. An apparatus for administering an asset-backed security, comprising:

a. a means for retrieving specifications of aspects of a plurality of assets, said assets generating cashflows from payment by obligors of principal and interest, wherein a level of the cashflows generated from said assets changes due to prepayments by said obligors;

b. a means for retrieving a specification of the asset-backed security created from said plurality of assets;

c. a means for determining, from said specifications of said aspects of the plurality of assets and said specification of said asset-backed security, a schedule of future cashflows that will be generated from the assets based on at least one of principal payments and interest payments from said obligors whose assets are members of said asset pool;

d. a means for modeling future prepayment behavior of said assets by said obligors, said modeling including the sub-steps of;

i. defining at least three refinancing efficiency categories of obligors according to an expected propensity of the obligors to refinance the assets,

ii. partitioning the assets in the asset pool into asset pool efficiency buckets according to said refinancing efficiency buckets,

iii. removing scheduled payments and core prepayments of principal which are not related to changing interest rates pro-rata from all asset pool efficiency buckets,

iv. attributing any excess of actual over scheduled prepayments to obligors refinancing assets in said asset pool, and

v. removing assets from said asset pool in a preset order according to the defined refinancing efficiency categories of obligors to reflect refinancing of assets by obligors and thus obtain a model of the remaining pool; and

e. a means for determining the current value of the asset-backed security, said current value determining step including the sub-steps of;

determining a core payment rate cash flow for said asset pool,

ii. generating a dual coupled interest rate lattice from an interest rate environment information including interest rates and respective interest rate volatilities, said lattice including a plurality of nodes representing points in time,

iii. determining, for the plurality of nodes in the in the lattice whether each defined refinancing category of said asset pool would reference based on the interest rate environment information,

iv. determining, for the plurality of nodes in the in the lattice, a respective first plurality of net cash flows for the asset-backed security, wherein the first plurality of net cash flows includes principal payments remaining in the asset pool after removal of assets due to refinancing,

v. determining, for the plurality of nodes in the in the lattice, a respective second plurality of net cash flows for the asset-backed security, the second plurality of net cash flows including both principal and interest payments, and

vi. determining a value for said asset-backed security by discounting and adding the first and second plurality of net cash flows determined for the plurality of nodes in the lattice.

9. The apparatus of claim 8, wherein said asset-backed securities are mortgage-backed securities and said obligors are mortgagors.

10. The apparatus of claim 8, wherein said asset-backed securities are collateralized mortgage obligations and said obligors are mortgagors.

11. The apparatus of claim 8, wherein said refinancing buckets of obligors include leapers who reference before an optimal time, financial engineers who reference at the optimal time, and laggards who reference after the optimal time.

12. The apparatus of claim 8, wherein said refinancing efficiency of obligors in deciding when to reference is determined by a savings/option value, said saving/option value being calculated assuming that interest rates for similar assets will follow a random walk.