Methods and systems for analytical-based multifactor Multiobjective portfolio risk optimization

• US 7,640,201 B2
• Filed: 03/19/2003
• Issued: 12/29/2009
• Est. Priority Date: 03/19/2003
• Status: Active Grant

• Current Assignee: GE Financial Assurance Holdings Incorporated (General Electric Company), General Electric Company
• Original Assignee: General Electric Company
• Inventors: Messmer, Richard Paul, Clark, Michael Craig, Chalermkraivuth, Kete Charles, Chakraborty, Anindya
• Primary Examiner(s): Havan; Thu-Thao
• Assistant Examiner(s): Hammond, III; Thomas M
• Application Number:
  US10/390,710
• Publication Number:
  US 20040199448A1
• Time in Patent Office: 2,477 Days
• Field of Search: 705 35- 40
• US Class Current: 705/36.00R
• CPC Class Codes: G06Q 40/06 Asset management; Financial planning or analysis

Abstract

The invention provides systems and methods for determining the allocation of securities in a portfolio. The method includes providing a collection of securities in a portfolio, each security being associated with associated attributes; providing risk factor data related to the portfolio; pooling the securities into a plurality of security clusters based on the attributes associated with each security and the risk factor data, each security being assigned to an security cluster, the pooling being performed using multivariate decision tree processing; processing the security clusters using a nonlinear programming optimizer to generate optimization results; and presenting the optimization results in a risk-return space for determination of a security allocation.

Claims

1. A method for determining the allocation of securities in a portfolio, the method comprising:

performing multivariate decision tree processing on a collection of securities in a portfolio using at least two concurrent target variables comprising a plurality of risk measures and at least one return measure, the plurality of risk measures including variance, skewness and Value at Risk, and pooling each security into a plurality of substantially homogeneous security clusters based on the target variables using a processor of a computing device;

determining pooled measures for each substantially homogeneous cluster, the pooled measures serving as proxies to each security in the portfolio, the pooled measures serving as inputs to the multivariate decision tree processing using the processor of the computing device;

processing the plurality of substantially homogeneous security clusters using a nonlinear programming optimizer to generate optimization results using the processor of the computing device; and

presenting the optimization results in a risk-return space for determination of a security allocation using the processor of the computing device.

2. The method of claim 1, wherein the pooled measures serve as constraints in the nonlinear programming optimizer.

3. The method of claim 1, further including decomposing the securities, before pooling the securities into a plurality of security clusters, into respective subsets of securities, and performing multivariate decision tree processing on each subset.

4. The method of claim 1, wherein the risk factor data is associated with risk measures, the method further including performing a risk measure simplification process though matrix manipulation comprising:

defining the change in risk factors;

defining Delta and Gamma;

restating the change in risk factors in Delta-Gamma formulation, the Delta-Gamma formulation having the factors Δ

F'"'"'s;

defining the covariance matrix of Δ

F;

taking the Cholesky decomposition of the covariance matrix to generate a P transformation matrix;

applying the P transformation matrix to Gamma to define a matrix Q_k;

determining the Eigenvalue decomposition of Q_k to obtain a matrix of Eigenvectors N; and

applying the matrix of Eigenvectors N and the P transformation matrix to evaluate the risk measures.

5. The method of claim 1, the portfolio is a collection of securities, which includes at least one of assets and liabilities.

6. A system for determining the allocation of securities in a portfolio, the system represented as a computer program tangibly embodied on a computer readable medium and executed by a computing device, comprising:

a security attribute portion, being at least one of an asset data portion and a liability data portion, containing attributes associated with a collection of securities in a portfolio;

a risk factor data portion that provides risk factor data related to securities in the portfolio;

a pooling portion that performs multivariate decision tree processing on the collection of securities in the portfolio using at least two target variables comprising a plurality of risk measures and at least one return measure the plurality of risk measures including variance, skewness and Value at Risk, pools each security into a plurality of substantially homogeneous security clusters based on the target variables, and determines pooled measures for each substantially homogeneous cluster, the pooled measures serving as proxies to each security in the portfolio, the pooled measures serving as inputs to the multivariate decision tree processing;

an optimization portion that processes the plurality of substantially homogeneous security clusters using a nonlinear programming optimizer to generate optimization results; and

a presentation portion that presents the optimization results in a risk-return space for determination of a security allocation.

7. The system of claim 6, wherein security attributes and the risk factor data provide a risk measure and a return measure for each security, the risk measure of each security and the return measure of each security being used as target variables in the multivariate decision tree processing.

8. The system of claim 7, wherein a first target is volatility and a second target is security return.

9. The system of claim 7, wherein the pooling portion decomposes the securities, before pooling the securities into a plurality of security clusters, into respective subsets of securities, each subset then being subject to multivariate decision tree processing.

10. An article of manufacturing for determining the allocation of securities in a portfolio, the article of manufacturing comprising:

a computer readable medium tangibly embodying a computer program executable by a processor of a computer, wherein the computer program contains;

a first portion containing attributes associated with a collection of securities in a portfolio;

a second portion that provides risk factor data related to securities in the portfolio;

a third portion that performs multivariate decision tree processing on the collection of securities in the portfolio using at least two concurrent target variables comprising a plurality of risk measures and at least one return measure the plurality of risk measures including variance, skewness and Value at Risk, pools each security into a plurality of substantially homogeneous security clusters based on the target variables, and determines pooled measures for each substantially homogeneous cluster, the pooled measures serving as proxies to each security in the portfolio, the pooled measures serving as inputs to the multivariate decision tree processing;

a fifth portion that processes the plurality of substantially homogeneous security clusters using a nonlinear programming optimizer to generate optimization results; and

a sixth portion that presents the optimization results in a risk-return space for determination of a security allocation.