estimate_curve: Perform the smoothing of individual curve.

In StevenGolovkine/denoisr: Nonparametric Kernel Smoothing Methods for Functional Data

Description

This function performs the smoothing of a curve using the Nadaraya-Watson estimator given a particular kernel.

Usage

estimate_curve(
  curve,
  U,
  b,
  t0_list = NULL,
  kernel = "epanechnikov",
  n_obs_min = 1
)

Arguments

curve	A list, with two entries: $t The sampling points $x The observed points.
U	A vector of numeric, sampling points at which estimate the curve.
b	Numeric or vector of numeric, estimation of the bandwidth. If one is provided, we use a unique bandwidth for the curve. However, if a vector is given, the bandwidth changes depending on the sampling points.
t0_list	A vector of numeric, times at which the bandwidths have been estimated. Only used if the parameter b is a vector.
kernel	Character string, the kernel used for the estimation: epanechnikov (default) uniform beta mBeta
n_obs_min	Integer, minimum number of observation for the smoothing

Value

A list, with two entries:

• $t The sampling points

• $x The estimated points.

Examples

X <- generate_piecewise_fractional_brownian(N = 1, M = 300, 
                                            H = c(0.2, 0.5, 0.8), 
                                            sigma = 0.05)
estimate_curve(X[[1]],
               U = seq(0, 1, length.out = 200),
               b = c(0.2, 0.5, 0.8),
               t0_list = c(0.16, 0.5, 0.83))

StevenGolovkine/denoisr documentation built on Nov. 15, 2021, 8:44 a.m.