lazy {RTisean}R Documentation

Nonlinear noise reduction

Description

Tools for performing nonlinear noise reduction.

Usage

nrlazy(series, l, x = 0, c = 1, m = 5, d = 1, i = 1, r, v)
lazy(series, m, r, v, i = 1, l, x = 0, c = 1)

Arguments

series a vector or a matrix.
m embedding dimension.
d delay for the embedding.
r neighborhood size.
v neighborhood size as fraction of data standard deviation.
i number of iterations.
l number of values to be read.
x number of values to be skipped.
c column to be read.

Details

In nrlazy each embedded point is replaced by the average vector calculated in its neighborhood with a given size. In lazy only the central component of each vector is corrected.

Value

A vector containing the filtered time series.

References

T. Schreiber, Extremely simple nonlinear noise reduction method, Phys. Rev. E 47, 2401 (1993)

See Also

ghkss,project

Examples

## Not run: 

par(mfrow=c(1,2))

n <- 5000
dat <- henon(n)
dat[,1] <- dat[,1] + rnorm(n)/25
delayeddat <- embed(dat,d=2)
filtereddat <- lazy(dat,m=5,v=0.05,c=2)
delayedfiltdat <- embed(filtereddat,d=2)
plot(delayeddat, cex= 0.2, cex.main=0.7, xlab="", ylab="", main = "Embedded noisy Henon attractor filtered with lazy")
points(delayedfiltdat, cex=0.2, col=2)
legend(-1,-0.5, c("Noisy data","Filtered data"),fill=c(1,2),bty="n",cex=0.7)

filteredat2 <- nrlazy(dat,m=5,v=0.2)[,1]
delayedfiltdat2 <- embed(filteredat2,d=2)
plot(delayeddat, cex= 0.2, cex.main=0.7,xlab="", ylab="", main =  "Embedded noisy Henon attractor filtered with nrlazy")
points(delayedfiltdat2 , cex=0.2, col=2)
legend(-1,-0.5, c("Noisy data","Filtered data"),fill=c(1,2),bty="n",cex=0.7)

## End(Not run)

[Package RTisean version 3.0.10 Index]