System and method for estimating a parameter that represents data describing a physical system

• US 8,290,880 B2
• Filed: 04/22/2009
• Issued: 10/16/2012
• Est. Priority Date: 04/22/2009
• Status: Expired due to Fees

• Current Assignee: Micro Focus LLC (Open Text Corporation)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Shan, Jerry Z., Safai, Fereydoon
• Primary Examiner(s): Zeender, Ryan
• Assistant Examiner(s): Chein, Allen
• Application Number:
  US12/427,924
• Publication Number:
  US 20100274642A1
• Time in Patent Office: 1,273 Days
• Field of Search: None
• US Class Current: 705/400
• CPC Class Codes: G06Q 30/0207 Discounts or incentives, e.g. coupons or rebates
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03

Abstract

There is provided a system and method for estimating a parameter that represents data describing a physical system. An exemplary method comprises randomizing data representative of a population of items for which the parameter is known. The method may additionally comprise generating data representative of a pseudo population of items using a known perturbation, the data representative of the pseudo population of items being included with the data representative of the population of items for which the parameter is known to form a revised population and selecting a bootstrap sample of a minimum sample size of the revised population. A sensitivity study is performed on the parameters of the items comprising the bootstrap sample to determine a level of change of a predicted parameter value relative to a parameter value of the sample. At least one of a range, a probability distribution or the minimum sample size is revised based on the parameter for items comprising the bootstrap sample to produce at least one of a revised range, a revised probability distribution or a revised minimum sample size, taking into account an effect of the known perturbation applied to the pseudo population. The steps of selecting, performing and revising are repeated until the sensitivity study indicates that the level of change of the parameter is acceptably small. A value of the parameter is estimated for the population based on a parameter corresponding to the acceptably small level of change.

Claims

1. A computer-implemented method for estimating via a processor a parameter that represents data describing a physical system, the method comprising:

randomizing data representative of a population of items for which the parameter is known, wherein the parameter corresponds to price consistency, and the physical system comprises at least one of a hardware element or a service element with respect to a networked information technology system;

generating data representative of a pseudo population of items using a known perturbation that provides a mean level shift of the parameter, the data representative of the pseudo population of items being included with the data representative of the population of items for which the parameter is known to form a revised population;

selecting a bootstrap sample of a minimum sample size of the revised population;

performing a sensitivity study on the parameter of the items comprising the bootstrap sample to determine a level of change of a predicted parameter value relative to a parameter value of the bootstrap sample;

revising at least one of a range, a probability distribution or the minimum sample size of a model that predicts the parameter based on the level of change to produce at least one of a revised range, a revised probability distribution or a revised minimum sample size, taking into account an effect of the known perturbation applied to the pseudo population;

repeating the steps of selecting, performing and revising until the sensitivity study indicates that the level of change of the predicted parameter value is acceptably small; and

estimating via the processor a value of the parameter for the population based on the predicted parameter value that corresponds to the acceptably small level of change.

2. The method recited in claim 1, wherein the known perturbation comprises random noise.

3. The method recited in claim 1, wherein the physical system comprises a financial system.

4. The method recited in claim 1, wherein the parameter corresponding to price consistency comprises a sales discount rate.

5. The method recited in claim 1, wherein the known perturbation comprises noise generated by a Gaussian process.

6. The method recited in claim 1, wherein the data representative of the pseudo population resembles aspects of the data representative of the original population but with the known perturbation reflecting uncertainty about the data representative of the population.

7. A computer system for estimating a parameter that represents data describing a physical system, the computer system comprising:

a processor that is adapted to execute stored instructions; and

a memory device that stores instructions that are executable by the processor, the instructions comprising;

computer-implemented instructions adapted to randomize data representative of a population of items for which the parameter is known, wherein the parameter corresponds to price consistency, and the physical system comprises at least a hardware element or a service element, or a combination thereof, regarding a networked information technology system;

computer-implemented instructions adapted to generate data representative of a pseudo population of items using a known perturbation comprising random noise generated by a Gaussian process, the data representative of the pseudo population of items being included with the data representative of the population of items for which the parameter is known to form a revised population;

computer-implemented instructions adapted to select a sample of a minimum sample size of the revised population;

computer-implemented instructions adapted to perform a sensitivity study on the parameter of the items comprising the sample to determine a level of change of a predicted parameter value relative to a parameter value of the sample;

computer-implemented instructions adapted to revise at least one of a range, a probability distribution or the minimum sample size of a model that predicts the parameter based on the level of change to produce at least one of a revised range, a revised probability distribution or a revised minimum sample size, taking into account an effect of the known perturbation applied to the pseudo population;

computer-implemented instructions adapted to repeat the steps of selecting, performing and revising until the sensitivity study indicates that the level of change of the predicted parameter value is acceptably small; and

computer-implemented instructions adapted to estimate a value of the parameter for the population based on the predicted parameter value that corresponds to the acceptably small level of change.

8. The computer system recited in claim 7, wherein the physical system comprises a financial system.

9. The computer system recited in claim 7, wherein the parameter corresponding to price consistency comprises a sales discount rate.

10. The computer system recited in claim 7, wherein the known perturbation provides a mean level shift of the parameter.

11. The computer system recited in claim 7, wherein the data representative of the pseudo population resembles aspects of the data representative of the original population but with the known perturbation reflecting uncertainty about the data representative of the population.