Systems and methods for forecasting based upon time series data
The present invention relates to systems and methods for forecasting using time series datasets. A composite may be generated by receiving datasets, normalizing them, and receiving formula configurations in order to combine the datasets together. The transformation of a dataset may be restricted if the accuracy of the transformation would be decreased, and if no suitable alternate dataset is available. A forecast may be generated using selected forecast type, calculation type, cutoff period, pre-adjustment, post-adjustment, indicators, and selected weights and offsets for the indicators. The forecast analysis may be updated by locking the time domain for one or more of the indicators. Forecast results may be outputted to a spreadsheet or other system utilizing add-ins. Any composite or forecast generated may be stored within a model repository for later use as an indicator.
1. A computerized method for generating a composite index, useful in association with a forecasting engine, the method comprising:
receiving selected datasets at a data aggregation server;
checking the selected datasets for relativity, wherein relativity is the degree of accuracy reduction of a dataset when converted for forecasting;
substituting at least one dataset above a relativity threshold when a substitute dataset is available;
rejecting at least one dataset above the relativity threshold when no substitute dataset is available;
normalizing the selected datasets;
receive formula configurations;
calculate composite by applying formula configurations to the normalized datasets by an application server of the forecasting engine; and
updating the calculated composite in real time as changes are made to the selected datasets.
2. The method of claim 1 wherein the selected dataset includes internal data and external data.
3. The method of claim 2 wherein the external data includes at least one of demographics data, meteorological data, climatic data, weather conditions, environmental data, industrial data and international financial market conditions.
4. The method of claim 1 wherein normalizing includes shifting a time domain for at least one of the selected datasets such that all of the normalized datasets have a common time domain.
5. The method of claim 1 wherein normalizing includes converting the selected datasets into a percentage value.
6. The method of claim 1 wherein normalizing includes converting the selected datasets into a dollar value.
7. The method of claim 1 wherein normalizing includes converting the selected datasets into a dollar versus time value.
8. A composite builder for generating a composite index, useful in association with a, the system comprising:
an aggregation server for receiving selected datasets; and
a logical engine for checking the selected datasets for relativity, wherein relativity is the degree of accuracy reduction of a dataset when converted for forecasting, substituting at least one dataset above a relativity threshold when a substitute dataset is available, rejecting at least one dataset above the relativity threshold when no substitute dataset is available, normalizing the selected datasets, receive formula configurations, calculate composite by applying formula configurations to the normalized datasets, and update the calculated composite in real time as changes are made to the selected datasets.
9. The system of claim 8 wherein the selected dataset includes internal data and external data.
10. The system of claim 9 wherein the external data includes at least one of demographics data, meteorological data, climatic data, weather conditions, environmental data, industrial data and international financial market conditions.
11. The system of claim 1 wherein normalizing includes shifting a time domain for at least one of the selected datasets such that all of the normalized datasets have a common time domain.
12. The system of claim 8 wherein normalizing includes converting the selected datasets into a percentage value.
13. The system of claim 8 wherein normalizing includes converting the selected datasets into a dollar value.
14. The system of claim 8 wherein normalizing includes converting the selected datasets into a dollar versus time value.