A statistical model that maximizes Lin's concordance correlation coefficient (CCC) instead of minimizing squared errors, offering a predictor with higher direct agreement with the predictand.