You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I had some questions
Q1] Is the SilverKite method supposed to take any regressor(like Xgboost) and convert that to an AutoRegressor(based on lags)?
Q2]Is the reconcile method meant to explore Hierarchical Forecasting as per Hyndman's?
It will be great if I can be any help creating example notebooks around the reconcile and Silverkite.
The text was updated successfully, but these errors were encountered:
To answer Q1:
Silverkite model accommodates both regular regressors and lagged regressors.
We have implemented a feature which creates "lagged regressors" and use that in the model. However that feature is NOT available in this release and should become available in the next one.
Please note that the current release does support regular regressors, in which case the values of those regressors are to be available in the future (as opposed to lagged regressors.)
The reconcile method imposes additive constraints (not necessarily hierarchical) on a set of forecasts, in a way that hopefully preserves or even improves forecast quality. It's a newer feature so we don't have a tutorial yet, but you are welcome to try it out. Our docstrings are fairly comprehensive, and you can check test cases for examples.
Hi, I had some questions
Q1] Is the SilverKite method supposed to take any regressor(like Xgboost) and convert that to an AutoRegressor(based on lags)?
Q2]Is the reconcile method meant to explore Hierarchical Forecasting as per Hyndman's?
It will be great if I can be any help creating example notebooks around the
reconcile
andSilverkite
.The text was updated successfully, but these errors were encountered: