Last data update: 2014.03.03

R: Smoothing with a finite difference penalty
difsmR Documentation

Smoothing with a finite difference penalty

Description

This function smoothes signals with a finite difference penalty of order 2.

Usage

difsm(y, lambda)

Arguments

y

signal to be smoothed: a vector

lambda

smoothing parameter: larger values lead to smoothing

Value

smoothed signal: a vector

Author(s)

Paul Eilers, Jan Gerretzen

References

Eilers, P.H.C. (2004) "Parametric Time Warping", Analytical Chemistry, 76 (2), 404 – 411.

Eilers, P.H.C. (2003) "A perfect smoother", Analytical Chemistry, 75, 3631 – 3636.

Examples

  data(gaschrom)
  plot(gaschrom[1,], type = "l", ylim = c(0, 100))
  lines(difsm(gaschrom[1,], lambda = 1e5), col = 2)
  lines(difsm(gaschrom[1,], lambda = 1e6), col = 3)
  lines(difsm(gaschrom[1,], lambda = 1e7), col = 4)
  

Results