Abstract
We investigate how to estimate error probabilities for binary measurements and tests when a gold standard is not available. Recent studies have shown that, without a gold standard, the widely used false-acceptance probability () and false-rejection probability (
) are difficult to estimate. We show by mathematical analysis that this problem is inherent to the unavailability of a gold standard. Instead, the proposed method determines the random components of the error probabilities: the inconsistent acceptance and rejection probabilities
and
. The method is efficient and robust as the number of model parameters is not predetermined but depends on the goodness of fit.
Additional information
Notes on contributors
Thomas S. Akkerhuis
Dr. Akkerhuis is a Statistical Consultant in the Statistics & Data Science team at Shell Global Solutions International. His e-mail address is [email protected].
Jeroen de Mast
Dr. De Mast is Professor in the Department of Statistics and Actuarial Science at the University of Waterloo. His email address is [email protected].
Tashi P. Erdmann
Dr. Erdmann is Manager HR Analytics at Shell International. His email address is [email protected].