The shunts I had made from ordinary copper seem to be much more stable with temperature.
Assuming that the temperature inside the meter tracks ambient pretty well, and assuming that there's no significant self heating of the shunt, I'd think it should track quite well as the tempco of the copper winding in the meter should match that of a copper wound shunt.
Run enough current through the shunt to warm it up , then its probably a different matter.
Ian VK3KRI