I am using the ss function from the npreg package in R to fit a smoothing cubic spline to my data, where the smoothing parameter is selected by the REML method. The "equivalent degrees of freedom" that is returned after fitting the model is different from the degrees of freedom listed in the model summary printout created after running the summary function on the model object. Screeenshot attached. Does anyone know why these degrees of freedom values are different from each other? One is 15.96, the other is 13.97.
