Abstract

Instructions for use of clinical chemical assays and other instruments commonly include figures for the precision at different levels of the measurand. Frequently neither the variance nor the coefficient of variation (CV) of the measure is constant across the range of measurand values, making such tables difficult to summarize or use. A more flexible variance model fills in the gap between the constant variance and constant CV models, leading to compact summaries of these precision tables and allowing interpolation to levels not included in the study. Likelihood ratio tests show when the more flexible model is needed.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call