Intelligent simulation analysis method and system

• US 7,865,415 B2
• Filed: 03/04/2004
• Issued: 01/04/2011
• Est. Priority Date: 03/05/2003
• Status: Active Grant

• Current Assignee: Morgan Stanley Services Group Inc. (Morgan Stanley)
• Original Assignee: Morgan Stanley
• Inventors: Cotton, Peter
• Primary Examiner(s): Borlinghaus; Jason M
• Application Number:
  US10/793,139
• Publication Number:
  US 20040236667A1
• Time in Patent Office: 2,497 Days
• Field of Search: None
• 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 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis
• CPC Class Codes: G06Q 50/12 Hotels or restaurants

Abstract

A method for calculating pricing information for a financial instrument consisting of a plurality of underlying financial instruments that includes the steps of: calculating a default time vector for a plurality of default scenarios wherein each default time vector includes a measure of a likelihood of default for each of the plurality of underlying financial instruments; calculating one or more cash flows for a subset of the default scenarios thereby forming a training set; training a neural network with the training set; and using the neural network to estimate one or more cash flows for a remaining number of the plurality of default scenarios.

Claims

1. A computer-implemented method for calculating pricing information of a financial instrument comprising a plurality of underlying financial instruments, the method comprising:

calculating in a computer system, a default time vector for each of a plurality of default scenarios wherein each default time vector includes a measure of a likelihood of default for each of said plurality of underlying financial instruments;

calculating in the computer system, one or more cash flows for a subset of said default scenarios;

training in the computer system, a neural network with said subset of said default scenarios, wherein the calculated cash flows for the subset of default scenarios are used as output training vectors for the neural network and the calculated default time vectors for each default scenario within the subset of default scenarios are used as input training vectors for the neural network; and

using said trained neural network in the computer system, to estimate one or more cash flows for a remaining number of said plurality of default scenarios.

2. The method of claim 1, further comprising the steps of:

determining in the computer system, a tranche impact parameter for one or more of said default scenarios in said subset of said default scenarios;

training in the computer system, a support vector machine with said tranche impact parameters associated with said default scenarios in said subset of said default scenarios; and

using a support vector machine in the computer system, to estimate a tranche impact parameter for one or more of said remaining number of said plurality of default scenarios;

wherein the step of training in the computer system, a neural network with said subset of said default scenarios includes the step of;

training in the computer system, a neural network with said tranche impact parameters associated with said default scenarios in said subset of said default scenarios; and

wherein the step of using said neural network in the computer system, to estimate one or more cash flows for a remaining number of said plurality of default scenarios includes the step of;

using said neural network in the computer system, to estimate one or more cash flows for said remaining number of said plurality of default scenarios based on said tranche impact parameters.

3. The method of claim 1, wherein the number of default scenarios comprising the subset is increased.

4. The method of claim 1, wherein the number of default scenarios comprising the subset is decreased.

5. The method of claim 2, wherein the tranche impact parameter for a default scenario is set to 1 when a level of default falls within a level of subordination associated with a tranche within the default scenario;

and the parameter is set to 0 when the level of default falls outside the level of subordination.

6. The method of claim 1, wherein said neural network is implemented in software.

7. The method of claim 1, wherein said neural network is implemented in hardware.

8. A computer system for calculating pricing information of a financial instrument comprising of a plurality of underlying financial instruments, the system comprising:

a programmable processor;

a computer software executable on the computer system;

a data storage system;

at least one input device; and

at least one output device;

the computer software operative with the processor to cause the data storage system to receive a plurality of default scenarios via the at least one input device; and

cause the processor to;

calculate a default time vector for each of the plurality of default scenarios wherein each default time vector includes a measure of a likelihood of default for each of said plurality of underlying financial instruments;

calculate one or more cash flows for a subset of said default scenarios;

train a neural network with said subset of said default scenarios, wherein the calculated cash flows for the subset of default scenarios are used as output training vectors for the neural network and the calculated default time vectors for each default scenario within the subset of default scenarios are used as input training vectors for the neural network;

use the trained neural network to estimate one or more cash flows for a remaining number of said plurality of default scenarios; and

forward said one or more cash flows to the at least one output device.

9. The computer system of claim 8, wherein the computer software is further operative with the processor to further cause the processor to:

determine a tranche impact parameter for one or more of said default scenarios in said subset of said default scenarios;

train a support vector machine with said tranche impact parameters associated with said default scenarios in said subset of said default scenarios;

use the support vector machine to estimate a tranche impact parameter for one or more of said remaining number of said plurality of default scenarios;

train the neural network with said tranche impact parameters associated with said default scenarios in said subset of said default scenarios; and

use said neural network to estimate one or more cash flows for said remaining number of said plurality of default scenarios based on said tranche impact parameters.

10. The system of claim 8, wherein the number of default scenarios comprising the subset is increased.

11. The system of claim 8, wherein the number of default scenarios comprising the subset is decreased.

12. The system of claim 9, wherein the tranche impact parameter for a default scenario is set to 1 when a level of default falls within a level of subordination associated with a tranche within the default scenario;

and the parameter is set to 0 when the level of default falls outside the level of subordination.

13. The system of claim 8, wherein said neural network is implemented in software.

14. The system of claim 8, wherein said neural network is implemented in hardware.