mescla.uncertainty.resampling

Jackknife and bootstrap uncertainty for mixing ratios.

Propagation and Monte Carlo both need you to state the end-member uncertainty. Resampling instead measures it, from the replicate samples that define each end-member. That is usually the honest route, because the dominant error is how well a handful of samples represents a source water – not how well the laboratory measured them.

This is the approach taken by recent EMMA implementations to attach a standard deviation to each sampling date.

Functions

bootstrap_endmembers	Bootstrap the replicates defining each end-member.
jackknife_endmembers	Leave-one-out uncertainty from the replicates defining each end-member.