Credibility Using a Loss Function from Spline Theory: Parametric Models with a One-Dimensional Sufficient Statistic [Author's Reply]