Systems and methods for forecasting household economics

• US 8,364,518 B1
• Filed: 02/14/2011
• Issued: 01/29/2013
• Est. Priority Date: 07/08/2009
• Status: Active Grant

• Current Assignee: Experian Limited (Experian plc)
• Original Assignee: Experian Limited (Experian plc)
• Inventors: Blake, Neil, Watts, Paul, McVittie, Eric
• Primary Examiner(s): LOFTIS, JOHNNA RONEE
• Application Number:
  US13/026,856
• Time in Patent Office: 715 Days
• Field of Search: 705/7.32
• US Class Current: 705/7.32
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Systems and methods of forecasting household economic data. In one embodiment, the forcast is made based on a macroeconomic scenario. In this aspect, a storage media of a computer maintains household survey records, and geodemographic segment identifiers are added to those records. A computer processor is used to aggregate the household survey records by geodemographic segments. The resulting segment-level records can be applied to an econometric model, and the model can use macroeconomic forecast assumptions related to the macroeconomic scenario, thus producing economic forecasts for the geodemographic segments. The economic forecasts that are produced may include predicted income and expenditures for each of the geodemographic segments.

Claims

1. A method of forecasting household economic data based on a macroeconomic scenario, the method comprising:

maintaining, in storage media of a computer, household survey records representing information about households;

adding, to the storage media, geodemographic segment identifiers to the household survey records, so that each of the household survey records is associated with a geodemographic segment;

aggregating, using a processor of the computer, the household survey records by the geodemographic segment identifiers to produce segment-level records; and

applying, using the processor, an econometric model to the segment-level records, the econometric model using macroeconomic forecast assumptions related to the macroeconomic scenario, the econometric model producing economic forecasts for each of the geodemographic segments;

wherein the produced economic forecasts include predicted income and expenditures for each of the geodemographic segments, the predicted income and expenditures relating to the macroeconomic scenario.

2. The method of claim 1, wherein adding the geodemographic segment identifiers to the household survey records comprises applying a geodemographic segmentation address lookup to the household survey records.

3. The method of claim 1, wherein adding the geodemographic segment identifiers to the household survey records comprises comparing characteristics of the household survey records to average characteristics associated with the geodemographic segments to determine matching geodemographic segments for the household survey records.

4. The method of claim 1, wherein the household survey records are grouped into household groups different from the geodemographic segments, and wherein adding the geodemographic segment identifiers to the household survey records comprises matching characteristics of the groups with characteristics associated with the geodemographic segments.

5. The method of claim 1, wherein applying the econometric model to the segment-level records comprises applying one or more macroeconomic models and/or macroeconomic satellite models.

6. The method of claim 1, wherein applying the econometric model to the segment-level records comprises applying one or more microeconometric and/or simulation models.

7. A computer system configured to construct an econometric model to forecast economic data for population segments, the system comprising:

an input data source maintaining historical economic data for individual population units;

a processor configured to communicate with the input data source;

an aggregation module, stored on a computer-readable medium and configured, when executed on the processor, to combine the historical economic data with geodemographic segmentation data comprising geodemographic segments corresponding to the individual population units; and

an analysis module, stored on the computer-readable medium and configured, when executed on the processor, to apply a statistical analysis procedure, using the historical economic data combined with the geographic segmentation data, to produce an econometric model configured to receive a set of macroeconomic variables and produce a forecast of economic data for each of the geodemographic segments.

8. The computer system of claim 7, wherein combining the historical economic data with the geodemographic segmentation data comprises applying a geodemographic segmentation address lookup to the historical economic data.

9. The computer system of claim 7, wherein combining the historical economic data with the geodemographic segmentation data comprises comparing characteristics of the historical economic data to characteristics associated with the geodemographic segmentation data to determine matching geodemographic segments for the historical economic data.

10. The computer system of claim 7, wherein the historical economic data is grouped into groups different from the geodemographic segments, and wherein combining the historical economic data with the geodemographic segmentation data comprises matching characteristics of the groups with characteristics associated with the geodemographic segments.

11. The computer system of claim 7, wherein applying the statistical analysis procedure comprises applying one or more macroeconomic models and/or macroeconomic satellite models.

12. The computer system of claim 7, wherein applying the statistical analysis procedure comprises applying one or more microeconometric and/or simulation models.

13. A method of predicting income and expenditures given macroeconomic situation assumptions, the method comprising:

maintaining an econometric model on a computer storage device, the econometric model being constructed with historical economic data for households mapped to population segment data, the model configured to predict income and expenditures for population segments based on macroeconomic situation assumptions;

applying, at a processor connected to the computer storage device, the econometric model to the given macroeconomic situation assumptions; and

receiving income and expenditure predictions for one or more population segments as output from the econometric model.

14. The method of claim 13, further comprising:

receiving household records belonging to a subgroup and associated with population segments;

identifying the income and expenditure predictions for the population segments associated with the household records; and

aggregating the identified income and expenditure predictions, to produce an aggregated income and expenditure prediction for the subgroup.

15. The method of claim 13, further comprising:

receiving a household profile comprising subgroups of households;

determining, for each subgroup, the fractions of the subgroup corresponding to the geodemographic segments; and

combining the income and expenditure predictions based on the determined fractions, to calculate subgroup income and expenditure predictions.

16. The method of claim 13, wherein the population segments are geodemographic household segments.