mjskay
Weighted quantiles collapse
Thanks a lot for your comparison of weighted quantile estimators! This has been very informative. Indeed there was a bug in the collapse algorithm (https://github.com/SebKrantz/collapse/issues/659) which I have just fixed (in version 2.0.17). The properties of the method assuming that the minimum weight is one unit appear reasonable.
I think though I'll implement your method if I can get it to run similarly fast. Your's is more theory consistent. I just tried to translate the h thresholds here to the weighted case.
collapse version 2.1.0 now implements your method of weighted quantiles (in C): https://sebkrantz.github.io/collapse/reference/fquantile.html
Many thanks again for the excellent note!