I'd like to run a piecewise cohort-sequential model. Two questions:
1. I haven't seen examples of piecewise cohort sequential models in the literature. In principle, is there any reason why one can't use the piecewise approach within a cohort-sequential model?
2. The model spans 43 years, so it's not obvious where the breakpoint should be. I could simply plot the data and "eyeball" it, but, is there a more precise way to determine the optimal location for a breakpoint?