Computer-implemented system and method for storing data analysis models

• US 8,010,324 B1
• Filed: 05/09/2006
• Issued: 08/30/2011
• Est. Priority Date: 05/09/2005
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Crowe, Keith E., Leonard, Michael J.
• Primary Examiner(s): FREJD, RUSSELL WARREN
• Application Number:
  US11/431,123
• Time in Patent Office: 1,939 Days
• Field of Search: 703/2, 703/22, 702/181, 702/189, 705/10, 707/104.1
• US Class Current: 703/2
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Computer-implemented systems and methods for processing time series data that is indicative of a data generation activity occurring over a period of time. A model specification hierarchical data structure is used for storing characteristics that define a time series model. A fitted model hierarchical data structure stores characteristics that define a fitted time series model. The characteristics of the fitted time series model are defined through an application of the time series model to the time series data.

Claims

1. A computer-implemented system for processing time series data that is indicative of a data generation activity occurring over a period of time, comprising:

a processor;

a computer-readable storage medium containing instructions configured to cause the processor to perform operations including;

storing characteristics that define a time series model in a model specification hierarchical data structure that includes a configuration for storing time series model characteristics in an Extensible Markup Language (XML) format;

storing characteristics that define a fitted time series model in a fitted model hierarchical data structure that includes a configuration for storing fitted time series model characteristics in an Extensible Markup Language (XML) format, wherein the fitted time series model characteristics include characteristics associated with a forecast function that describes how to forecast the time series data, and wherein the fitted time series model characteristics are defined through an application of the time series model to the time series data and

using the stored fitted time series model characteristics to provide forecasts of the time series data.

2. The system of claim 1, wherein the time series model characteristics include parameters optimized to fit the time series data.

3. The system of claim 1, wherein providing forecasts further comprises:

using a time series forecasting program to access the fitted time series model characteristics stored in the fitted model hierarchical data structure; and

using the time series forecasting program to generate a forecast based upon the accessed fitted time series model characteristics and based upon future input values.

4. The system of claim 3, wherein the time series forecasting program is configured to use the fitted time series model characteristics for goal seeking, optimization, scenario analysis, and control situations.

5. The system of claim 1, wherein the model specification hierarchical data structure is configured to store time series model characteristics independent of storage of the time series data.

6. The system of claim 1, further comprising instructions configured to cause the processor to perform operations including:

using forecast scoring generation software to generate and store the fitted time series model characteristics within the fitted model hierarchical data structure.

7. The system of claim 1, wherein the system further comprises one or more computer storage mediums for storing the model specification hierarchical data structure and the fitted model hierarchical data structure.

8. The system of claim 7, wherein the one or more computer storage mediums include mediums selected from RAM, ROM, CD-ROM, computer hard drives, and combinations thereof.

9. The system of claim 1, wherein the fitted time series model is used to generate backcasts.

10. The system of claim 1, wherein the fitted time series model is used to generate time series components.

11. The system of claim 10, wherein the time series components include seasonal components, and trend components.

12. The system of claim 1, further comprising instructions configured to cause the processor to perform operations including:

using forecast function generation software to generate a forecast function based upon a selected time series model, historical data and estimated parameters, wherein the forecast function is used to provide forecast scoring.

13. The system of claim 1, further comprising instructions configured to cause the processor to perform operations, including:

using model selection software to select one or more time series models to model the time series data.

14. A computer-implemented method for processing time series data that is indicative of a data generation activity occurring over a period of time, comprising:

storing, using one or more processors, characteristics that define a time series model in a model specification hierarchical data structure that includes a configuration for storing time series model characteristics in an Extensible Markup Language (XML) format;

storing, using one or more processors, characteristics that define a fitted time series model in a fitted model hierarchical data structure that includes a configuration for storing fitted time series model characteristics in an Extensible Markup Language (XML) format, wherein the fitted time series model characteristics include characteristics associated with a forecast function that describes how to forecast the time series data, and wherein the fitted time series model characteristics are defined through an application of the time series model to the time series data and

using the stored fitted time series model characteristics to provide forecasts of the time series data.

15. The method of claim 14, wherein the time series model characteristics include parameters optimized to fit the time series data.

16. The method of claim 14, wherein providing forecasts further comprises:

using a time series forecasting program to access the fitted time series model characteristics stored in the fitted model hierarchical data structure; and

using the time series forecasting program to generate a forecast based upon the accessed fitted time series model characteristics and based upon future input values.

17. The method of claim 16, wherein the time series forecasting program is configured to use the fitted time series model characteristics for goal seeking, optimization, scenario analysis, and control situations.

18. The method of claim 14, wherein the model specification hierarchical data structure is configured to store time series model characteristics independent of storage of the time series data.

19. The method of claim 14, further comprising:

using forecast scoring generation software to generate and store the fitted time series model characteristics within the fitted model hierarchical data structure.

20. The method of claim 14, wherein the fitted time series model is used to generate backcasts.

21. The method of claim 14, wherein the fitted time series model is used to generate time series components.

22. The method of claim 21, wherein the time series components include seasonal components, and trend components.

23. The method of claim 14, further comprising:

using forecast function generation software to generate a forecast function based upon a selected time series model, historical data and estimated parameters, wherein the forecast function is used to provide forecast scoring.

24. The method of claim 14, further comprising:

using model selection software to select one or more time series models to model the time series data.

25. Computer software stored on one or more computer-readable storage mediums, the computer software comprising program code for carrying out a method for processing time series data that is indicative of a data generation activity occurring over a period of time, the method comprising:

storing characteristics that define a time series model in a model specification hierarchical data structure that includes a configuration for storing time series model characteristics in an Extensible Markup Language (XML) format;

storing characteristics that define a fitted time series model in a fitted model hierarchical data structure that includes a configuration for storing fitted time series model characteristics in an Extensible Markup Language (XML) format, wherein the fitted time series model characteristics include characteristics associated with a forecast function that describes how to forecast the time series data, and wherein the fitted time series model characteristics are defined through an application of the time series model to the time series data and

using the stored fitted time series model characteristics to provide forecasts of the time series data.