iknots: Estimate internal knot positions equalizing jumps in n-th...

In bspline: B-Spline Interpolation and Regression

Estimate internal knot positions equalizing jumps in n-th derivative

Description

Normalized total variation of n-th finite differences is calculated for each column in y then averaged. These averaged values are smoothed (if requested) and sampled to find knot positions which equalize the jumps of n-th derivative. If any of found knots fall inside a same x interval, TV is smoothed till no such situation occurs.

Usage

iknots(x, y, nki = 1L, n = 3L, lenfit = 12L, smooth = FALSE)

Arguments

x	Numeric vector
y	Numeric vector or matrix
nki	Integer scalar, number of internal knots to estimate (1 by default)
n	Integer scalar, polynomial order of B-spline (3 by default)
lenfit	Integer scalar, length of knots for linear spline to fit the total variation if greater than 0, otherwise full linear interpolation is used (12 by default)
smooth	Logical scalar, should total variation be smoothed? (FALSE by default)

Value

Numeric vector, estimated knot positions

bspline documentation built on April 4, 2025, 4:54 a.m.