Nixtla / statsforecast

Lightning ⚡️ fast forecasting with statistical and econometric models.
https://nixtlaverse.nixtla.io/statsforecast
Apache License 2.0
3.97k stars 282 forks source link

ADAM model from R #886

Open ncooder opened 2 months ago

ncooder commented 2 months ago

Description

Are you going to add other models than AutoCES from smooth package like ADAM: here

Use case

No response

MMenchero commented 2 months ago

Hi @ncooder, Are there any specific models you would like to see included in StatsForecast? We are always striving to improve the library, but as you can imagine, adding new models is very time-consuming. So we prefer to prioritize features or models that our users need the most.

ncooder commented 2 months ago

@MMenchero Yes, there are more than AutoCES models that we can use in the statsforecast. The ADAM model is one of them from the smooth package that is great and probably no one will implement it in the Python besides the statsforecast. I can help with the development, but still I would need a guidance. I really appreciate the models you have there, but please try to add more SOTA models.

MMenchero commented 2 months ago

If you would like to implement the ADAM model, you can start with the contributing guide.

if you have specific questions, please let us know.

marcozanotti commented 1 month ago

Hi @MMenchero ,

related to the request there are not only a few SOTA methods like: