Rapid valuation of portfolios of assets such as financial instruments

• US 7,039,608 B2
• Filed: 12/14/2000
• Issued: 05/02/2006
• Est. Priority Date: 12/30/1999
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: GE Capital Commercial Finance Incorporated (General Electric Company)
• Inventors: Messmer, Richard Paul, Johnson, Christopher Donald, Akbay, Kunter Seref, Keyes, Tim Kerry, Midkiff, Catharine Lynn, Chen, Yu-to, Edgar, Marc Thomas, Pisupati, Chandrasekhar, Spencer, David Jonathan, Steward, William Cree, Rajiv, Vrinda, Cifarelli, James Louis, Nelson, David Richard
• Primary Examiner(s): Kazimi, Hani M.
• Assistant Examiner(s): CHARLES, DEBRA F
• Application Number:
  US09/737,454
• Publication Number:
  US 20020013752A1
• Time in Patent Office: 1,965 Days
• Field of Search: 705/40, 705/36, 705/35
• US Class Current: 705/36.00R
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• 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

Abstract

A method of valuation of large groups of assets by partial full underwriting, partial sample underwriting and inferred values of the remainder using an iterative and adaptive supervised and unsupervised statistical evaluation of all assets and statistical inferences drawn from the evaluation and applied to generate the inferred asset values. Individual asset values are developed and listed in realtional tables so that individual asset values can be rapidly taken from the tables and quickly grouped in any desired or prescribed manner for bidding purposes. The assets are collected into a database, divided into categories by credit variable, subdivided by ratings as to those variables and then rated individually. The assets are then regrouped according to a bidding grouping and a collective valuations established by cumulating the individual valuations.

Claims

1. A computer-implemented method for rapid valuation of asset portfolios using a portfolio valuation system, the portfolio valuation system including a computer coupled to a database, said method comprising the steps of:

valuating assets in a portfolio individually by segmenting the portfolio of assets into three valuation portions and by;

fully underwriting each asset included within a first portion of the asset portfolio for computing a value for each asset included within the first portion of the asset portfolio including underwriting in a full cash manner to generate a full value table, and underwriting in a partial cash manner to generate a partial value table,

storing in the database asset data acquired from fully underwriting the first portion including the computed value and descriptive attribute variables for each asset included within the first portion,

grouping and underwriting a sample of assets included within a second portion of the asset portfolio for valuation purposes based on the asset data acquired from the fully underwriting of the first portion and stored in the database, and

using the computer to statistically infer a value for each asset included within a third portion of the asset portfolio based on the asset data acquired from the fully underwriting of the first portion and stored in the database;

listing the asset values individually in relational tables;

aggregating to desired groups or tranches for bidding purposes; and

optimizing the bid pricing for desired risk/return tolerance.

2. A method according to claim 1 wherein said step of valuating assets further comprises the step of subjecting the assets in the portfolio to an iterative and adaptive valuation in which the assets in the portfolio are individually valued.

3. A method according to claim 1 wherein said step of grouping and underwriting a sample of assets included within a second portion of the asset portfolio further comprises the steps of:

using full sampling procedures for categories of large assets included within the portfolio; and

using partial sampling procedures for categories of small or medium assets included within the portfolio.

4. A method according to claim 3 wherein said step of using full sampling procedures comprises the steps of:

sampling one hundred percent of a sample group of assets; and

underwriting in full a portion of the sampling group of assets based on a determined commonality within the sampling group.

5. A method according to claim 4 further comprising the steps of:

generating a full sampling group valuation; and

desegregating the full sampling group valuation according to a rule set to produce an individual full sample asset value table.

6. A method according to claim 3 wherein said step of using partial sampling procedures comprises the steps of:

forming a cluster sample group;

sampling one hundred percent of a representative group from within the cluster;

randomly sampling other groups within the cluster; and

determining the values of the remaining group by extrapolating from the cluster sample group.

7. A method according to claim 6 further comprising the steps of:

re-underwriting at an asset level to produce an alpha credit analyst table;

asset class adjusting the alpha credit analyst table to produce an adjusted credit analyst table; and

selecting individual assets from the adjusted credit analyst table according to tranche grouping to produce a partial sampling credit value, for use in bidding.

8. A method according to claim 1 wherein said step of using the computer to statistically infer a value for each asset included within a third portion further comprises the step of utilizing both a supervised and an unsupervised learning process and a statistical inferencing algorithm to produce an underwriting clusters table which facilitates bid value.

9. A method according to claim 8 wherein the unsupervised leaming process further comprises the steps of:

segmenting and classifying assets; and

objectively evaluating the assets based upon underwriting or other valuation data feedback.

10. A method according to claim 7 wherein said step of using the computer to statistically infer a value for each asset included within a third portion of the asset portfolio comprises the step of clustering using fuzzy-C means clustering (FCM) and a composite High/Expected/Low/Timing/Risk (HELTR) scoring technique.

11. A method according to claim 10 further comprising the steps of:

segmenting into one category assets deemed to have sufficient commonality for evaluation as a whole; and

segmenting into a second category assets without sufficient commonality for evaluation as a whole.

12. A method according to claim 11 further comprising the steps of:

dividing the second category of assets into clusters; and

dividing the clusters into sub-clusters.

13. A method according to claim 12 further comprising the step of regrouping the sub-clusters into tranches for bidding purposes.

40. A method according to claim 1 wherein said step of valuating assets in a portfolio further comprises valuating assets in a portfolio wherein the assets include financial instruments.

14. A portfolio valuation system for rapid valuation of asset portfolios, said system comprising:

a computer configured as a server and further configured with a database of asset portfolios and to enable valuation process analytics;

at least one client system connected to said server through a network, said server configured to;

value assets in a portfolio individually;

segment the portfolio of assets into three valuation portions;

fully underwrite each asset included within a first portion of the asset portfolio for computing a value for each asset included within the first portion including underwriting in a full cash manner to generate a full value table, and underwriting in a partial cash manner to generate a partial value table;

store in the database asset data acciuired from fully underwriting the first portion including the computed value and descriptive attribute variables for each asset included within the first portion;

group and underwrite a sample of assets included within a second portion of the asset portfolio for valuation purposes based on the asset data acquired from the fully underwriting of the first portion and stored in the database;

statistically infer a value for each asset included within a third portion of the asset portfolio based on the asset data acciuired from the fully underwriting of the first portion and stored in the database;

list the asset values individually in tables;

aggregate to desired groups or tranches for bidding purposes; and

optimize the bid pricing for desired risk/return tolerance.

15. A system according to claim 14 wherein said server is configured to subject the assets in the portfolio to an iterative and adaptive valuation in which the assets in the portfolio are individually valued.

16. A system according to claim 14 wherein said server is configured to:

use full sampling procedures for categories of large assets included within the portfolio; and

use partial sampling procedures for categories of small or medium assets included within the portfolio.

17. A system according to claim 16 wherein said server is configured to:

sample one hundred percent of a sample group of assets; and

underwrite in full a portion of the sampling group of assets based on a determined commonality within the sampling group.

18. A system according to claim 17 wherein said server is configured to:

generate a full sampling group valuation; and

desegregate the full sampling group valuation according to a rule set to produce an individual full sample asset value table.

19. A system according to claim 16 wherein said server is configured to:

form a cluster sample group;

sample one hundred percent of a representative group from within the cluster;

randomly sample other groups within the cluster; and

determine the values of the remaining group by extrapolating from the cluster sample group.

20. A system according to claim 19 wherein said server is configured to:

re-underwrite at an asset level to produce an alpha credit analyst table;

asset class adjust the alpha credit analyst table to produce an adjusted credit analyst table; and

select individual assets from the adjusted credit analyst table according to tranche grouping to produce a partial sampling credit value, for use in bidding.

21. A system according to claim 14 wherein said server is configured to utilize both a supervised and an unsupervised learning process and an statistical inferencing algorithm to produce an underwriting clusters table.

22. A system according to claim 21 wherein said server is configured to:

segment and classify assets; and

evaluate the assets based upon data feedback.

23. A system according to claim 20 wherein said server is configured to cluster assets using fuzzy-C means clustering (FCM) and a composite High/Expected/Low/Timing/Risk (HELTR) scoring technique.

24. A system according to claim 23 wherein said server is configured to:

segment into one category assets deemed to have sufficient commonality for evaluation as a whole; and

segment into a second category assets without sufficient commonality for evaluation as a whole.

25. A system according to claim 24 wherein said server is configured to:

divide the second category of assets into clusters; and

divide the clusters into sub-clusters.

26. A system according to claim 25 wherein said server is configured to regroup the sub-clusters into tranches for bidding purposes.

41. A system according to claim 14 wherein the assets included within the portfolio include financial instruments.

27. A computer for rapid valuation of asset portfolios, said computer including a database of asset portfolios and configured to enable valuation process analytics, said computer programmed to:

value assets in a portfolio individually;

segment the portfolio of assets into three valuation portions;

fully underwrite each asset included within a first portion of the asset portfolio for computing a value for each asset included within the first portion including underwriting in a full cash manner to generate a full value table, and underwriting in a partial cash manner to generate a partial value table;

store in the database asset data acquired from fully underwriting the first portion including the computed value and descriptive attribute variables for each asset included within the first portion;

group and underwrite a sample of assets included within a second portion of the asset portfolio for valuation purposes based on the asset data acquired from the fully underwriting of the first portion and stored in the database;

statistically infer a value for each asset included within a third portion of the asset portfolio based on the asset data acquired from the fully underwriting of the first portion and stored in the database;

list the asset values individually in tables;

aggregate to desired groups or tranches for bidding purposes; and

optimize the bid pricing for desired risk/return tolerance.

28. A computer according to claim 27 programmed to subject the assets in the portfolio to an iterative and adaptive valuation in which the assets in the portfolio are individually valued.

29. A computer according to claim 27 programmed to:

use full sampling procedures for categories of large assets included within the portfolio; and

use partial sampling procedures for categories of small or medium assets included within the portfolio.

30. A computer according to claim 29 programmed to:

sample one hundred percent of a sample group of assets; and

underwrite in full a portion of the sampling group of assets based on a determined commonality within the sampling group.

31. A computer according to claim 30 programmed to:

generate a full sampling group valuation; and

desegregate the full sampling group valuation according to a rule set to produce an individual full sample asset value table.

32. A computer according to claim 29 programmed to:

form a cluster sample group;

sample one hundred percent of a representative group from within the cluster;

randomly sample other groups within the cluster; and

determine the values of the remaining group by extrapolating from the cluster sample group.

33. A computer according to claim 32 programmed to:

re-underwrite at an asset level to produce an alpha credit analyst table;

asset class adjust the alpha credit analyst table to produce an adjusted credit analyst table; and

select individual assets from the adjusted credit analyst table according to tranche grouping to produce a partial sampling credit value, for use in bidding.

34. A computer according to claim 27 programmed to utilize both a supervised and an unsupervised learning process and an statistical inferencing algorithm to produce an underwriting clusters table.

35. A computer according to claim 34 programmed to:

segment and classify assets; and

evaluate the assets based upon data feedback.

36. A computer according to claim 33 programmed to cluster assets using fuzzy-C means clustering (FCM) and a composite High/Expected/Low/Timing/Risk (HELTR) scoring technique.

37. A computer according to claim 36 programmed to:

segment into one category assets deemed to have sufficient commonality for evaluation as a whole; and

segment into a second category assets without sufficient commonality for evaluation as a whole.

38. A computer according to claim 37 programmed to:

divide the second category of assets into clusters; and

divide the clusters into sub-clusters.

39. A computer according to claim 38 programmed to regroup the sub-clusters into tranches for bidding purposes.

42. A computer according to claim 27 wherein the assets included within the portfolio include financial instruments.