Method and apparatus for an incomplete information model of credit risk

• US 7,536,329 B2
• Filed: 01/20/2004
• Issued: 05/19/2009
• Est. Priority Date: 01/17/2003
• Status: Expired due to Fees

• Current Assignee: Barra Incorporated
• Original Assignee: Barra Incorporated
• Inventors: Goldberg, Lisa Robin, Giesecke, Kay
• Primary Examiner(s): Cheung; Mary
• Assistant Examiner(s): FERTIG, BRIAN E
• Application Number:
  US10/761,686
• Publication Number:
  US 20040225598A1
• Time in Patent Office: 1,946 Days
• Field of Search: 705/35, 705/36.R, 705/38, 708/131, 708/132, 708/134, 434/107, 434/188, 434/191
• US Class Current: 705/35
• 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 40/08 Insurance

Abstract

A method and apparatus for developing a structural model of credit risk that incorporates the short-term uncertainty inherent in default events is disclosed. The model is based on the assumption of incomplete information, taking as premise that bond investors are not certain about the true level of a firm'"'"'s value that may trigger default. In addition, the coherent integration of structure and uncertainty is facilitated with compensators. Compensators form the infrastructure of a class of credit models that is broad enough to include traditional structural models, intensity-based models, and a great deal more. Several empirical examples are provided that compare default probabilities and credit yield spreads forecast by the incomplete information model to the output of a Black and Cox (1976) model. It is found that the incomplete information model reacts more quickly and, unlike traditional structural models, forecasts positive short-term credit spreads for firms that are in distress. It is also demonstrated that while the model is predicated on the surprise nature of default, it does not have conditional default rate.

Claims

1. A computer implemented method for generating a term-structure of default probabilities, comprising the steps of:

at least one computer determining a default process by performing the steps comprising;

determining a firm'"'"'s default barrier distribution from a plurality of risk factors comprising said firm'"'"'s value and an unobservable default barrier, said unobservable default barrier comprising an incomplete information model;

determining said firm'"'"'s conditional default probability over time using said default barrier distribution;

determining a pricing trend function using said conditional default probability where said pricing trend function estimates a probability of default of said firm; and

said at least one computer generating said term structure of default probabilities for said firm based on said pricing trend function.

2. The computer implemented method of claim 1, further comprising the step of:

calibrating parameters to represent the quality of incomplete information.

3. The computer implemented method of claim 1, further comprising the step of:

estimating diffusive and jump components of credit risk premium.

4. The computer implemented method of claim 1, further comprising the step of:

estimating market implied recovery rates.

5. The computer implemented method of claim 1, wherein said step of determining conditional default probability uses information comprising histories of equity prices, debt outstanding, agency ratings, and accounting variables.

6. The computer implemented method of claim 1, further comprising the steps of:

providing capability for triggering a default event when a firm value falls below a default barrier value;

providing capability for incorporating an assumption that said default barrier value is not publicly known;

providing capability for representing a predefault firm value process by a geometric Brownian motion; and

using a history of fundamental data and other publicly available information in determining said default barrier distribution and for estimating parameters of said firm value process.

7. The computer implemented method of claim 6, further comprising the steps of:

using histories of daily equity prices and equity volatility forecasts, reported liabilities, and risk-free interest rates as input to said step of determining a conditional default probability;

using option pricing formulae to convert said equity prices and said equity volatility forecasts into associated firm values and firm volatility;

estimating a mean and height of a scaled beta distribution from history of firm leverage ratios; and

providing capability for calibrating a degree of confidence about information by providing variance of said distribution as a free parameter.

8. The computer implemented method of claim 1, wherein said pricing trend function values credit-sensitive securities.

9. The computer implemented method of claim 1, further comprising:

said at least one computer creating fair values of credit-sensitive and default contingent securities based on said conditional default probability, said pricing trend, and said default barrier distribution.

10. The computer implemented method of claim 1, further comprising:

outputting to an investor, said term structure of default probabilities.

11. The computer implemented method of claim 1, further comprising:

22. The computer system of claim 8, further comprising:

determining a compensator using said conditional default process.

12. A computer system for generating a term structure of default probabilities comprising:

at least one computer memory having program code that when executed operates at least one computer to determine a default process by performing the steps comprising;

determining a firm'"'"'s default barrier distribution from a plurality of risk factors comprising said firm'"'"'s value and an unobservable default barrier, said unobservable default barrier comprising an incomplete information model;

determining said firm'"'"'s conditional default probability over time using said default barrier distribution;

determining a pricing trend function using said conditional default probability, where said pricing trend function estimates a probability of default of said firm; and

said at least one computer memory having program code that when executed operates said at least one computer to generate said term structure of default probabilities for said firm based on said pricing trend function.

13. The computer system of claim 12, further comprising:

calibrating parameters to represent the quality of incomplete information.

14. The computer system of claim 12, further comprising:

estimating diffusive and jump components of credit risk premium.

15. The computer system of claim 12, further comprising:

estimating market implied recovery rates.

16. The computer system of claim 12, wherein said step of determining conditional default probability uses information comprising histories of equity prices, debt outstanding, agency ratings, and accounting variables.

17. The computer system of claim 12, further comprising:

capability for triggering a default event when a firm value falls below a default barrier value

capability for incorporating an assumption that said default barrier value is not publicly known;

capability for representing a predefault firm value process by a geometric Brownian motion; and

using a history of fundamental data and other publicly available information in determining said default barrier distribution and for estimating parameters of said firm value process.

18. The computer system of claim 17, further comprising:

using daily equity prices and equity volatility forecasts, reported liabilities, and risk-free interest rates as input to said step of determining a conditional default probability;

using option pricing formulae to convert said equity prices and said equity volatility forecasts into associated firm values and firm volatility;

estimating a mean and height of a scaled beta distribution from history of firm leverage ratios; and

calibrating degree of confidence about information by providing variance of said distribution as a free parameter.

19. The computer system of claim 12, wherein said pricing trend function values credit-sensitive securities.

20. The computer system of claim 12, further comprising:

creating fair values of credit-sensitive and default contingent securities based on said conditional default probability, said pricing trend, and said default barrier distribution.

21. The computer system of claim 12, further comprising:

said at least one computer memory having program code that when executed operates at least one computer to output to an investor said term structure of default probabilities.

23. A computer program product comprising a computer useable medium having control logic stored therein for causing at least one computer to generate a term structure of default probabilities comprising:

computer readable program code for causing said at least one computer to determine a default process by performing the steps comprising;

determining a firm'"'"'s default barrier distribution from a plurality of risk factors comprising said firm'"'"'s value and an unobservable default barrier, said unobservable default barrier comprising an incomplete information model;

determining said firm'"'"'s conditional default probability over time using said default barrier distribution;

determining a pricing trend using said conditional default probability, where said pricing trend function estimates a probability of default of said firm; and

computer readable program code for causing said at least one computer to generate said term structure of default probabilities for said firm based on said pricing trend function.

24. The computer program product of claim 23, further comprising:

causing said at least one computer to calibrate parameters to represent the quality of incomplete information.

25. The computer program product of claim 23, further comprising:

causing said at least one computer to estimate diffusive and jump components of credit risk premium.

26. The computer program product of claim 23, further comprising:

causing said at least one computer to estimate market implied rates.

27. The computer program product of claim 23, wherein said step of determining conditional default probability uses information comprising histories of equity prices, debt outstanding, agency ratings, and accounting variables.

28. The computer program product of claim 23, further comprising:

causing said at least one computer to provide capability for triggering a default event when a firm value falls below a default barrier value;

causing said at least one computer to provide capability for incorporating an assumption that said default barrier value is not publicly known;

causing said at least one computer to provide capability for representing a predefault firm value process by a geometric Brownian motion; and

causing said at least one computer to use a history of fundamental data and other publicly available information in determining said default barrier distribution and for estimating parameters of said firm value process.

29. The computer program product of claim 28, further comprising:

causing said at least one computer to use histories of daily equity prices and equity volatility forecasts, reported liabilities, and risk-free interest rates as input to said step of determining said conditional default probability;

causing said at least one computer to use option pricing formulae to convert said equity prices and said equity volatility forecasts into associated firm values and firm volatility;

causing said at least one computer to estimate a mean and height of a scaled beta distribution from history of firm leverage ratios; and

causing said at least one computer to provide capability for calibrating a degree of confidence about information by providing variance of said distribution as a free parameter.

30. The computer program product of claim 23, wherein said pricing trend function values credit-sensitive securities.

31. The computer program product of claim 23, further comprising:

creating fair values of credit-sensitive and default contingent securities based on said conditional default probability, said pricing trend, and said default barrier distribution.

32. The computer program product of claim 23, further comprising:

computer readable program code for causing said at least one computer to output to an investor said term structure of default probabilities.

33. The computer program product of claim 23, further comprising:

causing said at least one computer to determine a compensator using said conditional default process.