Apply existing model to other time series
- Dominant language
- Python
- Stars
- 20.4k
- Forks
- 4.6k
- Avg merge
- 19h 52m
- Merged PRs (30d)
- 1
Description
Hi,
How can I use existing fitted models (I've created) to create forecasts for other sets of data?
Example:
Say I have 10k items to create models for, time constraints don't allow me to create a model for every item, so then I will instead create a model for its Product Family (that are only 100), in which 99% of its items follow the same trend/seasonality.
Now how can I run the item level time series through the product group family fitted models to extract item level forecasts? Which should basically mimic the product group family seasonality/trend by of course applied to that item's specific values.
Something equivalent to what I'm asking would be AutoARIMA.forward
Contributor guide
Assessment
This issue has not been assessed yet.