formalize links between location effects and changes in variance #23
Description
when non-biological samples were added to the model, it made me consider how to model other correlated effects of 'transitions' from a free-living to host-associated lifestyle, and vice versa. For instance, it could be that a host-associated lifestyle increases or decreases rate of evolution. So, a single parameter could be useful that assesses such a linkage: an increase in location effect for 'isBiological' has a corresponding increase in variance effect, perhaps related via a single global scaling parameter?