Hello all! Is there any way to predict at a different cutoff than the one at the end of my training set? For instance, MLForecast allows the

df_new

in predict, is there something similar in statsforecast?

Hey. Some models implement the

forward

method, but not all of them and I don't think we have a method in the

StatsForecast

class. Feel free to open an issue if it'd be valuable to you. It's not that common in statsforecast because there's a single model per serie, so transfer learning isn't really a thing but it may perform ok for the same serie at a future time