Robust estimators of the concordance correlation coefficient.

This paper proposes a generalized version of Lin' s concordance correlation coefficient for the agreement assessment of continuous data. Lin's coefficient evaluates the accuracy and precision between two measures, and is based on the expected value of the squared distance function. We generalize Lin's coefficient, apply alternative distance functions, and produce more robust versions of the concordance correlation coefficient. In this paper, we develop the asymptotic theory for this class of estimators, investigate small-sample properties via computer simulation, and demonstrate their use with two real data examples.