Changes from a default of regressing by counterpart_area to counterpart_type to reduce the noise (smoother data, more linear) Changes the window to start in 2020 because COVID represents a structural break.
@lpicci96 FYI this changes the PR you reviewed. I'll go for a merge because of time, but we should unpack/discuss on Monday.
Changes from a default of regressing by counterpart_area to counterpart_type to reduce the noise (smoother data, more linear) Changes the window to start in 2020 because COVID represents a structural break.
@lpicci96 FYI this changes the PR you reviewed. I'll go for a merge because of time, but we should unpack/discuss on Monday.
@sharcourt14 FYI