Computer-implemented systems and methods for forecast reconciliation

• US 8,364,517 B2
• Filed: 12/16/2011
• Issued: 01/29/2013
• Est. Priority Date: 11/03/2006
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Trovero, Michele Angelo, Joshi, Mahesh V., Leonard, Michael James, Fahey, Richard Patrick, Golovashkin, Dmitry V.
• Primary Examiner(s): MARCUS, LELAND R
• Application Number:
  US13/328,811
• Publication Number:
  US 20120089609A1
• Time in Patent Office: 410 Days
• Field of Search: 705/7
• US Class Current: 705/7.31
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 10/063 Operations research, analysis or 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

Systems and methods for reconciling a forecast are presented. A method can be used that receives a plurality of hierarchical forecast data sets. An output child data set including an index value and a status indicator representing an unprocessed state is generated. A particular parent data set forecast is identified from a parent data set. Locations for a group of one or more child data set forecasts that are children of the particular parent data set forecast are identified and accessed. A reconciliation operation is performed, a particular child data set forecast is adjusted and stored in a record, and a status indicator for the record is modified.

Claims

1. A computer-implemented method for reconciling hierarchical forecasts, comprising:

receiving, using one or more processing units, a plurality of input hierarchical forecast data sets, wherein the hierarchical forecast data sets include forecasts independently generated for each of a plurality of nodes hierarchically arranged according to parent child relationships, wherein one or more of the forecasts are included in records of a parent data set, wherein one or more of the forecasts are included in records of a child data set, and wherein the records of the child data set are in a particular order according to an index;

generating, the one or more processing units, an output child data set including one record for each of the records in the child data set in an identical order as the child data set, wherein the output child data set record includes an index value and a status indicator representing an unprocessed state;

identifying, using the one or more processing units, a particular parent data set forecast from the parent data set;

identifying, using the one or more processing units, locations for a group of one or more child data set forecasts that are children of the particular parent data set forecast;

accessing, using the one or more processing units, the one or more identified child data set forecasts using the identified locations;

performing, using the one or more processing units, a reconciliation operation based upon the particular parent data set forecast and the one or more accessed child data set forecasts, wherein the reconciled particular parent data set forecast is mathematically equivalent to the aggregation of the reconciled one or more accessed child data set forecasts, wherein the reconciled forecasts include associated measures of uncertainty, and wherein the reconciliation operation adjusts a particular child data set forecast;

storing, using the one or more processing units, the adjusted child data set forecast in a record of the output child data set using the index value corresponding to the particular child data forecast; and

changing, the one or more processing units, the status indicator for the record of the output child data set to represent a processed state.

2. The method of claim 1, wherein the parent data set records include one or more parent entity attributes, wherein the child data set records include one or more child entity attributes, and wherein the index values for the group of one or more child data set forecasts are identified based upon a match between the parent entity attributes and the child entity attributes.

3. The method of claim 1, wherein the adjustment to the particular child data set forecast results in a sum of the group of child set forecasts equaling the particular parent data set forecast.

4. The method of claim 1, wherein the adjustment to the particular child data set forecast results in an average of the group of child set forecasts equaling the particular parent data set forecast.

5. The method of claim 1, further comprising:

identifying a next parent data set forecast from the parent data set;

identifying further locations for a group of one or more further child data set forecasts that are children of the next parent data set forecast;

accessing the one or more identified further child data set forecasts using the identified further locations;

performing a reconciliation operation based upon the next parent data set forecast and the one or more accessed further child data set forecasts, wherein the reconciled next parent data set forecast is mathematically equivalent to the aggregation of the reconciled one or more accessed further child data set forecasts, wherein the reconciled forecasts include associated measures of uncertainty, and wherein the reconciliation operation adjusts a particular child data set forecast;

storing the adjusted further child data set forecast in a further record of the output child data set using an index value corresponding to the particular further child data forecast; and

changing the status indicator for the further record of the output child data set to represent a processed state.

6. The method of claim 1, further comprising:

storing copies of a plurality of the records of the child data set in memory, wherein locations for the group of one or more child data set forecasts that are children of the particular parent data set forecast are identified using the stored copies.

7. The method of claim 6, further comprising:

replacing one or more of the stored copies with one or more additional copies of records of the child data set.

8. The method of claim 6, wherein a number of copies stored is based upon an amount of memory space designated as cache storage.

9. The method of claim 6, further comprising:

storing a pointer to a first record in the plurality of records in the child data set, wherein the first record is not stored in the memory.

10. The method of claim 9, further comprising:

using the pointer to replace one or more of the stored copies with one or more additional copies of records in the child data set.

11. A computer-implemented system, comprising:

one or more processors;

one or more non-transitory computer-readable storage media containing instructions configured to cause the one or more processors to perform operations including;

receiving a plurality of input hierarchical forecast data sets, wherein the hierarchical forecast data sets include forecasts independently generated for each of a plurality of nodes hierarchically arranged according to parent child relationships, wherein one or more of the forecasts are included in records of a parent data set, wherein one or more of the forecasts are included in records of a child data set, and wherein the records of the child data set are in a particular order according to an index;

generating an output child data set including one record for each of the records in the child data set in an identical order as the child data set, wherein the output child data set record includes an index value and a status indicator representing an unprocessed state;

identifying a particular parent data set forecast from the parent data set;

identifying locations for a group of one or more child data set forecasts that are children of the particular parent data set forecast;

accessing the one or more identified child data set forecasts using the identified locations;

performing a reconciliation operation based upon the particular parent data set forecast and the one or more accessed child data set forecasts, wherein the reconciled particular parent data set forecast is mathematically equivalent to the aggregation of the reconciled one or more accessed child data set forecasts, wherein the reconciled forecasts include associated measures of uncertainty, and wherein the reconciliation operation adjusts a particular child data set forecast;

storing the adjusted child data set forecast in a record of the output child data set using the index value corresponding to the particular child data forecast; and

changing the status indicator for the record of the output child data set to represent a processed state.

12. The system of claim 11, wherein the parent data set records include one or more parent entity attributes, wherein the child data set records include one or more child entity attributes, and wherein the index values for the group of one or more child data set forecasts are identified based upon a match between the parent entity attributes and the child entity attributes.

13. The system of claim 11, wherein the adjustment to the particular child data set forecast results in a sum of the group of child set forecasts equaling the particular parent data set forecast.

14. The system of claim 11, wherein the adjustment to the particular child data set forecast results in an average of the group of child set forecasts equaling the particular parent data set forecast.

15. The system of claim 11, further comprising instructions configured to cause the one or more processors to perform operations including:

identifying a next parent data set forecast from the parent data set;

identifying further locations for a group of one or more further child data set forecasts that are children of the next parent data set forecast;

accessing the one or more identified further child data set forecasts using the identified further locations;

performing a reconciliation operation based upon the next parent data set forecast and the one or more accessed further child data set forecasts, wherein the reconciled next parent data set forecast is mathematically equivalent to the aggregation of the reconciled one or more accessed further child data set forecasts, wherein the reconciled forecasts include associated measures of uncertainty, and wherein the reconciliation operation adjusts a particular child data set forecast;

storing the adjusted further child data set forecast in a further record of the output child data set using an index value corresponding to the particular further child data forecast; and

changing the status indicator for the further record of the output child data set to represent a processed state.

16. The system of claim 11, further comprising instructions configured to cause the one or more processors to perform operations including:

storing copies of a plurality of the records of the child data set in memory, wherein locations for the group of one or more child data set forecasts that are children of the particular parent data set forecast are identified using the stored copies.

17. The system of claim 16, further comprising instructions configured to cause the one or more processors to perform operations including:

replacing one or more of the stored copies with one or more additional copies of records of the child data set.

18. The system of claim 16, wherein a number of copies stored is based upon an amount of memory space designated as cache storage.

19. The system of claim 16, further comprising instructions configured to cause the one or more processors to perform operations including:

storing a pointer to a first record in the plurality of records in the child data set, wherein the first record is not stored in the memory.

20. The system of claim 19, further comprising instructions configured to cause the one or more processors to perform operations including:

using the pointer to replace one or more of the stored copies with one or more additional copies of records in the child data set.

21. A computer-program product tangibly embodied in a non-transitory computer-readable storage medium, including instructions configured to cause a data processing system to:

receive a plurality of input hierarchical forecast data sets, wherein the hierarchical forecast data sets include forecasts independently generated for each of a plurality of nodes hierarchically arranged according to parent child relationships, wherein one or more of the forecasts are included in records of a parent data set, wherein one or more of the forecasts are included in records of a child data set, and wherein the records of the child data set are in a particular order according to an index;

generate an output child data set including one record for each of the records in the child data set in an identical order as the child data set, wherein the output child data set record includes an index value and a status indicator representing an unprocessed state;

identify a particular parent data set forecast from the parent data set;

identify locations for a group of one or more child data set forecasts that are children of the particular parent data set forecast;

access the one or more identified child data set forecasts using the identified locations;

perform a reconciliation operation based upon the particular parent data set forecast and the one or more accessed child data set forecasts, wherein the reconciled particular parent data set forecast is mathematically equivalent to the aggregation of the reconciled one or more accessed child data set forecasts, wherein the reconciled forecasts include associated measures of uncertainty, and wherein the reconciliation operation adjusts a particular child data set forecast;

store the adjusted child data set forecast in a record of the output child data set using the index value corresponding to the particular child data forecast; and

change the status indicator for the record of the output child data set to represent a processed state.

22. The computer-program product of claim 21, wherein the parent data set records include one or more parent entity attributes, wherein the child data set records include one or more child entity attributes, and wherein the index values for the group of one or more child data set forecasts are identified based upon a match between the parent entity attributes and the child entity attributes.

23. The computer-program product of claim 21, wherein the adjustment to the particular child data set forecast results in a sum of the group of child set forecasts equaling the particular parent data set forecast.

24. The computer-program product of claim 21, wherein the adjustment to the particular child data set forecast results in an average of the group of child set forecasts equaling the particular parent data set forecast.

25. The computer-program product of claim 21, further comprising instructions configured to cause a data processing system to:

identify a next parent data set forecast from the parent data set;

identify further locations for a group of one or more further child data set forecasts that are children of the next parent data set forecast;

access the one or more identified further child data set forecasts using the identified further locations;

perform a reconciliation operation based upon the next parent data set forecast and the one or more accessed further child data set forecasts, wherein the reconciled next parent data set forecast is mathematically equivalent to the aggregation of the reconciled one or more accessed further child data set forecasts, wherein the reconciled forecasts include associated measures of uncertainty, and wherein the reconciliation operation adjusts a particular child data set forecast;

store the adjusted further child data set forecast in a further record of the output child data set using an index value corresponding to the particular further child data forecast; and

change the status indicator for the further record of the output child data set to represent a processed state.

26. The computer-program product of claim 21, further comprising instructions configured to cause a data processing system to:

store copies of a plurality of the records of the child data set in memory, wherein locations for the group of one or more child data set forecasts that are children of the particular parent data set forecast are identified using the stored copies.

27. The computer-program product of claim 26, further comprising instructions configured to cause a data processing system to:

replace one or more of the stored copies with one or more additional copies of records of the child data set.

28. The computer-program product of claim 26, wherein a number of copies stored is based upon an amount of memory space designated as cache storage.

29. The computer-program product of claim 26, further comprising instructions configured to cause a data processing system to:

store a pointer to a first record in the plurality of records in the child data set, wherein the first record is not stored in the memory.

30. The computer-program product of claim 29, further comprising instructions configured to cause a data processing system to:

use the pointer to replace one or more of the stored copies with one or more additional copies of records in the child data set.