Skip to contents

Smoothing spline

Source: R/smoothing.R
tda_spl.Rd

A smoothing spline through a pair of variables, optionally weighted. smooth.spline is the R analogue.

Usage

tda_spl(
  x,
  y,
  weights = NULL,
  sig = 1,
  deg = NULL,
  max = 200,
  rx = NULL,
  options = list(),
  dir = tempfile("tda")
)

Arguments

x, y

the pair of variables.

weights

optional per-point weights, the same idea as weights on the fitting functions.

sig

smoothing factor; 0 interpolates.

deg

degree of the spline.

max

maximum number of knots.

rx

range over which to evaluate, e.g. seq(0, 10, 1).

options

a named list of further TDA options, passed through.

dir

working directory.

Value

An object carrying a table of fitted values.

See also

Other smoothing: tda_integrate(), tda_interp(), tda_isotonic(), tda_mat(), tda_sma(), tda_smd()

Examples

set.seed(33)
x <- sort(runif(30, 0, 10))
y <- sin(x) + rnorm(30, sd = 0.15)
sp <- tda_spl(x, y, sig = 1)
head(sp$table)
#>   index         x         y    fitted          d1         d2
#> 1     1 0.1551696 0.1522848 0.2769707  0.88128812 -0.6805919
#> 2     2 0.4273416 0.3929474 0.4916244  0.69605006 -0.6805919
#> 3     3 1.1799116 0.9726723 0.8227202  0.18385699 -0.6805919
#> 4     4 1.3576507 0.9957205 0.8446484  0.06288921 -0.6805919
#> 5     5 2.2505121 0.6885517 0.6295154 -0.54478500 -0.6805919
#> 6     6 2.6048568 0.4449743 0.3937460 -0.78594920 -0.6805919