Age Regrouping

[sbodenstein]

This feature regroups any count data for arbitrary age ranges using the R ungroup package. For example,

regroup_counts_pclm(['0-2', '3-4'], [2,5], max_age=5)
# output (['0-0', '1-1', '2-2', '3-3', '4-4'], [0.41, 0.81, 1.44, 2.09, 2.21])
regroup_counts_pclm(['0-2', '3-6', '7+'], [3,9,12], new_ages=['0-1', '2-5', '6+'])
# output (['0-1', '2-5', '6+'], [2.01, 7.4, 16.9])

Some design questions:

• the method comes with confidence intervals as well, but not currently returning them. Are they useful?
• currently, new_ages is optional. If its None (default), it uses year-increments. Might make sense to make this non-optional, and force user to give new age ranges.
• The output is a tuple currently. Might make sense to rather give it as a dictionary of pandas dataframe
• the new counts from the method are not integers. Should this function do the rounding? Note that the rounding can violate a basic consistency criteria (the total counts should remain invariant post re-grouping)