search

fhui28 / CBFM

Spatio-temporal joint species distribution modeling using community-level basis functions
13 stars 2 forks source link

Effective degrees of freedom #5

Closed fhui28 closed 3 years ago

fhui28 commented 3 years ago

Maybe necessarily for smooths and significant tests? Need to check what mgcv does...

fhui28 commented 3 years ago

Decided to only do this smoothing terms in formula_X, and not the spatial and/or temporal BF terms [since I do not really want to be necessarily advocating that sparse variable selection really be done on the BFs...although this could be something considered down the road].

As a result, the issue of edf is now pulled into the summary.CBFM issue #1