NEWS.md
coef to the plot() method for splines2 objects, allowing visualization of the fitted spline function with a given coefficient vector.nsk() for natural cubic spline basis functions following the function survival::nsk() (introduced in survival package version 3.2-8).plot() methods to quickly visualize the spline basis functions.$ method to extract an attribute of the returned splines2 object.periodic to bSpline() for periodic B-splines and a new class named PeriodicBSpline to the Rcpp interface: issue 19.coef to the predict() methods to compute the responding spline function and made it possible to obtain the derivatives or update spline basis functions by passing ... to the update() methods.trim to naturalSpline() to set the default boundary knots after trimming a fraction of observations.warn.outside and a package option named splines2.warn.outside to specify if a warning should be thrown out for B-splines, etc. when any x is placed outside the boundary.H to the attribution of objects for natural cubic splines so that users may transform cubic B-splines (from other software/packages) to the natural cubic splines (returned by naturalSpline()/nsp() or nsk()).
PeriodicMSpline so that a simple knot sequence can be specified through set_knot_sequence: issue 18.update() methods to produce new spline basis functions based on the given object with specified updates in terms of degree and knots, etc.makepredictcall() methods for all available spline basis functions to help model.frame.default() create the right matrices when predicting from models with terms such as bSpline(), etc. Thanks Zheyuan Li for suggesting this feature.derivs and integal to bSpline() for consistency with mSpline() and bernsteinPoly(), etc.predict() method for cSpline objects when scale = FALSE.knots() methods to extract internal knots and boundary knots from a given splines2 object.naturalSpline() providing implementation of nonnegative natural cubic splines.periodic to function mSpline() for periodic M-splines.integral to function mSpline() for integrals of M-splines or periodic M-splines.deriv(), predict(), and print() method for naturalSpline class object.deriv() method for mSpline class object for periodic M-splines.bernsteinPoly() providing implementation of generalized Bernstein polynomials.intercept in function iSpline() and cSpline() to TRUE for a complete set of spline basis functions in shape-restricted regression.iSpline() and cSpline().bSpline().dbs() generating derivative of given order of B-splines. It is a similar function with splines::splineDesign(). However, it provides a more user-friendly interface and more consistent handling on NA’s.deriv() methods for derivatives of given order of any existing splines2 object that can be generated currently.bSpline() generating B-spline basis allowing zero degree or piecewise constant basis based on function bs() in the splines package.bSpline() to allow M-splines of degree zero.cSpline() constructing convex spline (C-spline) basis.predict() methods for bSpline2 object and cSpline object generated by bSpline() and cSpline(), respectively.print() methods for all splines2 objects developed so far.iSpline() to construct I-spline basis directly from B-spline basis instead of M-spline basis.