dtwSat (version 0.1.0)

waveletSmoothing: Wavelet filter

Description

This function performs a smoothing algorithm to the time series. It computes a discreat wavelet smoothing for each dimension in the imput time series.

Usage

waveletSmoothing(x, timeline = NULL, frequency = NULL, wf = "la8", J = 1, boundary = "periodic", ...)

Arguments

x

A zoo object with the time series

timeline

A vector of dates for the output time series. It must have a regular frequency.

frequency

The frequncy for the output time series

wf

Name of the wavelet filter used in the decomposition. Default is "la8"

J

Specifies the depth of the decomposition. This must be a number less than or equal to log(length(x),2). Default is 1

boundary

Character string specifying the boundary condition. Default is "periodic". See parameters of mra.

...

see parameters of mra in the packege waveslim

Value

object of class zoo

See Also

mra

Examples

## Wavelet filter
sy = waveletSmoothing(x=template, frequency=16, wf = "la8", J=1,
     boundary = "periodic")
plot(template$evi, xlab="Time", ylab="EVI")
lines(sy$evi, col="red")

## Plot raw EVI and filtered EVI
#require(ggplot2)
#evi = merge(Raw=zoo(template$evi), Wavelet=zoo(sy$evi))
#gp = autoplot(evi, facets = NULL) + xlab("Time")
#gp

## Plot all filter bands
#require(ggplot2)
#gp = autoplot(sy, facets = NULL) + xlab("Time")
#gp