Portfolio optimization and evaluation tool

• US 10,664,914 B2
• Filed: 07/21/2014
• Issued: 05/26/2020
• Est. Priority Date: 07/21/2014
• Status: Active Grant

• Current Assignee: Validus Holdings Ltd. (American International Group Incorporated)
• Original Assignee: American International Group Incorporated
• Inventors: Hofmann, Georg Wilhelm
• Primary Examiner(s): Jacob, William J
• Application Number:
  US14/336,632
• Publication Number:
  US 20160019647A1
• Time in Patent Office: 2,136 Days
• 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 computer system configured to evaluate a portfolio comprising instruments, comprising a computer memory configured to store, for each instrument, an instrument value for each portfolio scenario in an n-dimensional matrix, a first constraint and a second constraint; and a computer processor configured to transpose the n-dimensional matrix, to determine a first solution by maximizing the product of transpose of the n-dimensional matrix and the first constraint, determine whether the first solution is within an accepted range of an acceptable risk, if the expected first solution is not within an accepted range of an acceptable risk, process the second constraint with the first solution to obtain a second solution, and determine whether the second solution is within the accepted range of the acceptable risk.

Claims

1. A computer system comprising:

a processor; and

a tangible, non-transitory memory configured to communicate with the processor,

the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations comprising;

storing, by the processor, a plurality of portfolio scenarios and, for each instrument, an instrument value for each of the plurality of portfolio scenarios in an n-dimensional matrix, a first constraint and a second constraint;

transposing, by the processor, the n-dimensional matrix, to determine a first solution by maximizing the product of a transpose of the n-dimensional matrix and the first constraint;

determining, by the processor, a conditional value at risk (CVaR);

determining, by the processor, an accepted range of an acceptable risk based on the CVaR,

determining, by the processor, whether the first solution is within the accepted range of the acceptable risks;

if the first solution is not within the accepted range of the acceptable risk, processing, by the processor, the second constraint with the first solution to obtain a second solution; and

if the first solution is not within the accepted range of the acceptable risk, determining, by the processor, whether the second solution is within the accepted range of the acceptable risk.

2. The computer system of claim 1, wherein the first constraint is a first level of risk.

3. The computer system of claim 2, wherein the second constraint is a second level of risk.

4. The computer system of claim 3, wherein the first constraint is a profit margin.

5. The computer system of claim 1, wherein the portfolio scenarios comprise insurance events and the severity for each of the insurance events.

6. The computer system of claim 1, further comprising displaying the first solution if the first solution is within the accepted range of acceptable risk or the second solution if the second solution is within the accepted range of acceptable risk.

13. The computer system of claim 1, wherein the determining the CVaR is by calculating a risk vector for each portfolio scenario.

15. The computing system of claim 1, wherein the first constraint comprises one of a transaction cost balance constraint, a value constraint, and a liquidity constraint.

7. A method comprising:

transposing, by a processor, an n-dimensional matrix that comprises a portfolio of instruments and an instrument value for each of a plurality of portfolio scenarios that is saved in computer memory;

determining, by the processor, a first solution by maximizing a product of the transpose of the n-dimensional matrix and a first constraint;

determining, by the processor, a conditional value at risk (CVaR) by calculating a risk vector for each portfolio scenario;

determining, by the processor, an accepted range of an acceptable risk based on the CVaR;

determining, by the processor, whether the first solution is within the accepted range of the acceptable risk and outputting the first solution to a computer system display if the first solution is within the accepted range of acceptable risk;

if the first solution is not within the accepted range of the acceptable risk, processing, by the processor, a second constraint with the first solution to obtain a second solution; and

if the first solution is not within the accepted range of the acceptable risk, determining, by the processor, whether the second solution is within the accepted range of the acceptable risk.

8. The method of claim 7, wherein the first constraint is a first level of risk.

9. The method of claim 8, wherein the second constraint is a second level of risk.

10. The method of claim 7, wherein the first constraint is a profit margin.

11. The method of claim 7, wherein the portfolio of scenarios comprise insurance events.

12. The method of claim 11, wherein the portfolio of scenarios further comprise severity for each event.

14. The method of claim 7, further comprising displaying the first solution or the second solution on a computer system display.

16. The method of claim 7, wherein the first constraint comprises one of a transaction cost balance constraint, a value constraint, and a liquidity constraint.