Searching for a package that can make predictions at unseen locations using multiscale geographically weighted regression

[nikos_geo]

I am searching for some time now for a package that can perform prediction at unseen locations using multiscale geographically weighted regression (MGWR). By unseen locations I mean at a finer spatial scale.

I am aware of the GWmodel package and the function gwr.multiscale, but I can't use it for prediction (it's not supported).

Does anyone knows if such function/package exists in R? I have seen few papers the used MGWR for prediction (e.g., Spatial Downscaling of NPP-VIIRS Nighttime Light Data Using Multiscale Geographically Weighted Regression and Multi-Source Variables, and other) but it's been over a week now and I can't find such package.

[nirgrahamuk]

when you says its not supported, can you explain that ?

[nikos_geo]

The function gwr.multiscale can be used for prediction at the same spatial resolution as the input data but cannot be used to predict data at unseen locations (e.g., fine spatial scale).