Possibly there is a better way to break ties in lx than the proposed maybe_dither_lx() better would be only to perturb lx in the very ages where ties are observed, but I didn't think through how it would work. Anyway, this solves some problems in the quantile functions for now
Possibly there is a better way to break ties in
lxthan the proposedmaybe_dither_lx()better would be only to perturblxin the very ages where ties are observed, but I didn't think through how it would work. Anyway, this solves some problems in the quantile functions for now