An approach using extreme value theory (EVT) to generate conservative overbounds for measurement errors is assessed. In this approach, EVT is used to construct a model for tails of an unknown distribution. Results from a Monte Carlo simulation study are presented, which show that estimated tails are not necessarily conservative. The reasons for this lack of conservatism are described and discussed. A method for addressing this lack of conservatism is proposed and evaluated.
No takes yet. Share an insight, caveat, or question.
Larson et al. (2017) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: