This package is for fitting free-knot splines for data with one independent variable and one dependent variable.
Four fitting methods are included for the case where the number of knots is known in advance; for details, see
freeknotfit. In addition, methods are available to compute the fitted values, the residuals, and the coefficients of the splines, and to plot the results, along with a method to summarize the results. Finally, a function (see
fit.search.numknots) is provided to optimize the number of knots using a specified fit criterion. Several fit criteria are provided (see
Steven Spiriti, Philip Smith, and Pierre Lecuyer
Spiriti, S., Eubank, R., Smith, P., Young, D., "Knot Selection for Least-Squares and Penalized Splines," Journal of Statistical Computation and Simulation, in press.