Looking at these graphs, it seems to me that the multiple modes are driven by age more than cohort. It looks like the first peak is dominated by age 1s, the second peak dominated by ages 2 and 3, and then the third peak dominated by the rest of the age groups.
Does that align with what you see? Should I plot it a different way?
color by age (limited to first 7 ages)
color by cohort
Looking at these graphs, it seems to me that the multiple modes are driven by age more than cohort. It looks like the first peak is dominated by age 1s, the second peak dominated by ages 2 and 3, and then the third peak dominated by the rest of the age groups.
Does that align with what you see? Should I plot it a different way?