Rsquared for higher hierarchical levels

[elildojr]

The Rsquared function apparently only works for the sampling unit level. I've been trying to calculate R2 for higher hierarchical levels by hand, using the instructions on page 12 of the HMSC manual (summing sampling-unit level predictions, comparing them with observed abundances etc, still trying) but it would be cool if there was a function with arguments allowing to choose hierarchical level.